MESOZOIC RIFTING AND CENOZOIC BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES Inferences from tectonic models
VRIJE UNIVERSITEIT
MESOZOIC RIFTING AND CENOZOIC BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES Inferences from tectonic models
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus Prof.dr. T. Sminia, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de faculteit der Aard- en Levenswetenschappen op dinsdag 19 maart 2002 om 13.45 uur in het hoofdgebouw van de universiteit, De Boelelaan 1105
door Luis Fernando Sarmiento Rojas
geboren te Santafé de Bogotá Colombia
promotor: copromotor:
prof.dr. S.A.P.L. Cloetingh dr. F. Roure
Luis Fernando Sarmiento, 2001 Copyright Ecopetrol 2001 © Vrije Universiteit Amsterdam ©
The research reported in this thesis was carried out at the Tectonics Department Faculty of Earth and Life Sciences, Vrije Universiteit Amsterdam De Boelelaan 1085 1081 HV Amsterdam The Netherlands Netherlands Research School of Sedimentary Geology (NSG) Publication number: 2002.01.01 Financial support was provided by ECOPETROL Empresa Colombiana de Petróleos And the ECOPETROL-ICETEX fund. ISBN 9287-25-5
Si un hombre nunca se contradice, será porque nunca dice nada Miguel de Unamuno. (If a man never contradicts himself, it will be because he never says anything)
To my wife Gladytas My mother, my sons Cesar and Daniel and to the memory of my father Filiberto
CONTENTS
ACKNOWLEDGEMENTS
xii
SUMMARY
xiv
SAMENVATTING
xx
RESUMEN
xxvi 1
CHAPTER 1: INTRODUCTION AND OUTLINE
1.1 1.2 1.3 1.4 1.5
AIM AND METHODS BASIN FORMATION AND INVERSION MODELS THE EASTERN CORDILLERA OF COLOMBIA TECTONIC SETTING OF THE EASTERN CORDILLERA THESIS OUTLINE
CHAPTER 2: MESOZOIC RIFTING HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES
1. INTRODUCTION 2. TECTONIC SETTING 2.1 PLATE-TECTONIC INTERPRETATIONS 2.1.1 Triassic and Jurassic 2.1.2 Cretaceous 3. STRATIGRAPHY 3.1 TRIASSIC AND JURASSIC SYN-RIFT SEDIMENTATION 3.2 CRETACEOUS SEDIMENTATION 3.2.1 Early Cretaceous Cretaceous Syn-Rift Sedimentation 3.2.2 Cretaceous Post-Rift Sedimentation 4. SUBSIDENCE ANALYSIS 4.1 TECTONIC SUBSIDENCE DURING TRIASSIC AND JURASSIC TIME 4.1.1 Basin Compartments 4.1.2 Triassic and Jurassic Subsidence Events 4.2 TECTONIC SUBSIDENCE DURING CRETACEOUS TIME 4.2.1 Basin Compartments 4.2.2 Cretaceous Fast Subsidence Events 5. CORRELATION OF FAST SUBSIDENCE EVENTS WITH MAGMATIC, EUSTATIC AND PLATE-TECTONIC EVENTS 5.1 CORRELATION BETWEEN FAST SUBSIDENCE EVENTS AND SUBDUCTION RELATED MAGMATIC ARCS 5.2 CORRELATION BETWEEN FAST SUBSIDENCE EVENTS, PLATE-TECTONIC EVENTS AND EUSTATIC EVENTS 6. FORWARD MODELLING OF BASIN EVOLUTION 6.1 NUMERICAL MODEL 6.2 MODELLING PROCEDURE 6.3 MODEL RESULTS: STRETCHING FACTORS
vii
1 1 2 2 3
7
7 8 8 8 12 12 12 18 18 26 31 34 37 37 42 42 44 48 48 52 54 54 55 56
6.3.1 Mesozoic Lithosphere Stretching Phases 6.3.2 Total Stretching 7. DISCUSSION 7.1 GEOMETRY OF RIFT BASINS 7.2 RELATIONSHIPS BETWEEN MESOZOIC RIFTING AND MAGMATISM 7.3 SUGGESTIONS FOR FUTURE STUDIES 8. CONCLUSIONS
CHAPTER 3: PALAEOGENE INCIPIENT BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES
1. INTRODUCTION 2. TECTONIC SETTING 2.1 PLATE TECTONIC INTERPRETIONS 2.1.1 Latest Cretaceous and Tertiary 3. STRATIGRAPHY 3.1 LATE MAASTRICHTIAN-EARLY PALEOCENE 3.2 LATE PALEOCENE 3.3 PALAEOGENE (EOCENE TO EARLY MIOCENE) 3.3.1 Llanos and Eastern Cordillera 3.3.2 Magdalena Valley (MV) 4. SUBSIDENCE ANALYSIS OF THE PALAEOGENE SEDIMENTARY RECORD 4.1. RESULTS 4.1.1 Events of Tectonic Subsidence 4.1.2 Maps of Tectonic Subsidence 5. MODELLING OF REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING 5.1 RESULTS 6. FLEXURAL BEHAVIOUR OF THE LITHOSPHERE 6.1 MODEL DESCRIPTION 7. MODELLING OF FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PALAEO-CENTRAL CORDILLERA DURING PALAEOGENE TIME 7.1 2D FLEXURAL MODELLING 7.2 3D FLEXURAL MODELLING 7.3 RESULTS 7.3.1 2D Models 7.3.2 3D Models 8. MODELLING OF FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PARTIALLY INVERTED EXTENSIONAL BASIN DURING PALAEOGENE TIME 8.1 RESULTS 9. MODELLING OF SUBSIDENCE PRODUCED BY THE COMBINED EFFECT OF (A) REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING, (B) FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PALAEO-CENTRAL CORDILLERA AND (C) FLEXURAL SUBSIDENCE PRODUCED BY LOCAL TOPOGRAPHY DUE TO PARTIAL INVERSION OF THE MESOZOIC EXTENSIONAL BASIN 9.1 2D MODELS 9.1.1 Results 9.2 3D MODEL 10. COMPARISON OF MODEL RESULTS WITH FISSION TRACK DATA AND OTHER EVIDENCE OF UPLIFT AND DEFORMATION 10.1 FISSION TRACK DATA AND OTHER EVIDENCE OF EXHUMATION AND DEFORMATION 10.2. COMPARISON WITH MODEL RESULTS 11. STRUCTURAL KINEMATIC MODEL OF PALAEOGENE DEFORMATION, INCLUDING SEDIMENTATION AND EROSION EVENTS 11.1 STRUCTURAL, SEDIMENTARY AND EROSION MODEL 11.2 RESULTS 12. DISCUSION 12.1 CONSTRAINING DIFFERENT TECTONIC SCENARIOS FOR THE PALAEOGENE
viii
56 61 62 62 66 70 70
72
72 74 74 74 76 76 78 80 81 84 93 93 94 96 97 98 99 101 103 103 105 105 105 107 108 109
109 109 111 117 119 119 124 124 124 125 128 128
12.2 PALAEOGENE TECTONIC HISTORY AND PLATE-TECTONICS 12.3 MECHANICAL ASPECTS OF BASIN INVERSION 12.3.1 Stresses 12.3.2 Rheology of the Lithosphere 12.3.3 Pre-Existing Crustal Discontinuities 12.4 LARGE-SCALE BASIN WIDE FLEXURE AND STRUCTURAL GEOLOGY 12.5 SUGGESTIONS FOR FUTURE STUDIES 13. CONCLUSIONS
CHAPTER 4: NEOGENE BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES
1. INTRODUCTION 2. TECTONIC SETTING 2.1 PLATE TECTONIC INTERPRETATIONS 2.1.1 Middle Miocene 2.1.2 Late Miocene–Pliocene 3. STRATIGRAPHY 3.1 LLANOS ORIENTALES (LLA) 3.2 MAGDALENA VALLEY (MV) 3.2.1 Middle Magdalena Valley 3.2.2 Upper Magdalena Valley 3.3 SABANA DE BOGOTÁ 4. TECTONIC SUBSIDENCE DURING THE NEOGENE 4.1 RESULTS 4.1.1 Llanos Orientales (LLA) 4.1.2 Magdalena Valley (MV) 5. NEOGENE REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING 5.1 RESULTS 6. NEOGENE FLEXURAL SUBSIDENCE 6.1. FLEXURAL MODELS PRODUCED BY THE PRESENT-DAY TOPOGRAPHIC LOAD OF THE EASTERN AND CENTRAL CORDILLERA 6.1.1 Methods 6.1.2 Results 6.2 FLEXURAL MODELS PRODUCED BY THE PRESENT-DAY TOPOGRAPHIC LOAD OF THE EASTERN CORDILLERA ONLY 6.2.1 Flexural Deflection of the Lithosphere as a Mechanism of Tectonic Subsidence During Neogene Time 6.2.2 Calculated Bending Stress 6.3 FLEXURAL SUBSIDENCE MODELS WITH A BROKEN PLATE UNDER THE EASTERN CORDILLERA 6.3.1 Methods 6.3.2 Results 6.4 LATERAL DISTRIBUTION OF EFFECTIVE ELASTIC THICKNESS AT PRESENT 7. FLEXURAL SUBSIDENCE PRODUCED BY GRADUAL SURFACE-UPLIFT OF THE EASTERN CORDILLERA DURING NEOGENE TIME 7.1 DATA 7.2 METHOD 7.3 RESULTS 8. UPLIFT EVOLUTION FROM FLEXURAL MODELING, FISSION TRACK AND GEOLOGICAL DATA 8.1 SURFACE-UPLIFT EVOLUTION OF THE EASTERN CORDILLERA AS INFERRED FROM FLEXURAL MODELLING 8.2 COMPARISON OF THE UPLIFT AND EXHUMATION EVOLUTION INFERRED FROM FISSION TRACK AND GEOLOGICAL DATA 9. DISCUSSION 9.1 NEOGENE TECTONIC HISTORY AND PLATE-TECTONICS 9.2 LITHOSPHERE STRENGTH EVOLUTION AS INFERRED FROM FLEXURAL MODELLING
ix
129 130 130 130 132 132 133 133 135
135 135 135 135 135 137 137 140 142 142 145 145 146 146 146 146 147 148 149 149 150 151 151 155 155 155 156 157 158 159 160 161 162 162 167 171 171 172
9.2.1 Effect of Lithosphere Thermal Age on Strength 9.2.2 Coupling/Decoupling State of the Crust-Mantle, Thickness and Proportions of Mechanically Competent Crust and Mantle 9.2.3 Effect of Local Curvature of the Plate on Lithosphere Strength 9.2.4 Effect of Pre-Existing Discontinuities on Lithosphere Strength 9.2.5 Other Effects on Lithosphere Strength 9.2.6 Temporal Changes of Lithosphere Strength 9.2.7 Suggestions for Further Studies 10. CONCLUSIONS
CHAPTER 5: MAP VIEW RESTAURATION OF TRANSPRESSIONAL BASIN INVERSION IN THE EASTERN CORDILLERA
1. INTRODUCTION 2. TECTONIC SETTING 2.1 PRESENT DAY PLATE-TECTONIC SETTING 3. STRUCTURE OF THE EASTERN CORDILLERA 4. REGIONAL MAP VIEW RESTORATION OF THE NW CORNER OF SOUTH AMERICA 4.1 METHOD 4.2 RESULTS 5. MAP VIEW RESTORATION OF THE EASTERN CORDILLERA 5.1 METHOD 5.2 RESULTS 5.3 DISPLACEMENT OF BLOCKS RELATIVE TO STABLE SOUTH AMERICA AND ROTATIONS ABOUT VERTICAL AXES 6. DISCUSION 6.1 ADVANTAGES OF THE MAP VIEW RESTORATION 6.2 COMPARISON WITH PREVIOUS SHORTENING ESTIMATES 6.3 COMPARISON WITH OUTCROP STRUCTURAL STUDIES, STRESS INFERRED FORM BOREHOLE BREAKOUT DATA AND PLATE MOTIONS FROM GEOPHYSICAL DATA 6.3.1 Shortening Perpendicular to the Regional Structural Grain of the EC 6.3.2 Conjugate (?) Left-lateral and right-lateral strike-slip faults 6.3.3 Clockwise rotation of the Central Cordillera, Magdalena Valley and Western Flank of the Eastern Cordillera 6.3.4 Deformation of the Eastern Cordillera by Transpressive Basin Inversion 6.3.5 Minor Structures Recognized in Outcrops, but not in the Map View Restoration 6.4 COMPARISON OF DISPLACEMENT OF BLOCKS RELATIVE TO STABLE SOUTH AMERICA WITH PLATE VELOCITY VECTORS FROM GEOPHYSICAL DATA 6.5 ANDEAN NEOGENE DEFORMATION, BASIN INVERSION OF THE EASTERN CORDILLERA AND LITHOSPHERE RHEOLOGY 7. CONCLUSIONS
CHAPTER 6: RHEOLOGICAL EVOLUTION OF THE LITHOSPHERE OF THE EASTERN CORDILLERA AND HYPOTHESES ABOUT ITS deep STRUCTURE
1. INTRODUCTION 2. GEOPHYSICAL DATA CONSTRAINING THE DEEP STRUCTURE OF THE EASTERN CORDILLERA 2.1 SEISMIC VELOCITY MODEL BASED ON REFRACTION IN THE SOUTH WEST OF COLOMBIA 2.2 GRAVITY AND MOHO DISCONTINUITY DEPTH 2.3 SEISMICITY 2.4 GEOTHERMAL REGIME
x
174 175 177 178 178 178 179 179
181
181 181 181 184 190 190 191 192 192 197 201 201 201 201 203 203 204 211 212 213 213 215 215
217
217 217 217 217 220 227
3. RHEOLOGICAL EVOLUTION OF THE LITHOSPHERE OF THE EASTERN CORDILLERA 3.1 RHEOLOGY OF THE CONTINENTAL LITHOSPHERE 3.1.1 Rheology of the Continental Lithosphere from Rock Mechanics Data 3.1.2 Rheological Profiles and Integrated Strength of a Stratified Lithosphere 3.1.3 Thermal Structure of Continental Lithosphere 3.2 RHEOLOGY MODELS OF THE EASTERN CORDILLERA THROUGH TIME 3.2.1 Rheology Models 3.2.2 Local Isostasy vs Regional Isostasy Effects 3.2.3 Model Results 3.2.4 Comparison of Lithosphere Strength Estimates form Rheological Models and Flexural Models 3.2.5 Models of Evolution of Lithosphere Rheology and Basin Extension and Inversion in the EC 4. EVOLUTION OF STRESSES AFFECTING THE LITHOSPHERE OF THE EASTERN CORDILLERA 4.1 MESOZOIC 4.2 CENOZOIC 5. MESOZOIC-CENOZOIC TECTONIC EVOLUTION OF THE EASTERN CORDILLERA 5.1 MESOZOIC EXTENSIONAL BASIN FORMATION 5.2 CENOZOIC TRANSPRESSIONAL INVERSION OF MESOZOIC EXTENSIONAL BASINS 6. COMPARISON OF THE EASTERN CORDILLERA WITH SIMILAR MONTAIN BELTS, ANALOGUE AND NUMERICAL MODELLING EXPERIMENTS AND HYPOTHESES ABOUT THE DEEP STRUCTURE OF THE EASTERN CORDILLERA 6.1 ASYMMETRY OF THE EASTERN CORDILLERA 6.2 COMPARISON AND SIMILARITIES WITH THE PYRENEES AND THE MÉRIDA ANDES OF VENEZUELA 6.3 FLEXURAL MODELLING OF BROKEN PLATE AND SUBDUCTION OF THE MANTLE LITHOSPHERE 6.4 COMPARISON WITH ANALOGUE MODEL EXPERIMENTS 6.5 COMPARISON WITH NUMERICAL MODELS OF BEAUMONT ET AL. (2000) AND ELLIS AND BEAUMONT (1999) 6.5.1 Model Features 6.5.2 Results 6.6 NEOGENE VOLCANISM AND DEEP-INTERMEDIATE SEISMICITY: TWO ENIGMATIC FEATURES OF THE EASTERN CORDILLERA 6.6.1 Neogene Magmatism of Paipa Iza 6.6.2 The Bucaramanga Earthquake Nest 6.6.3 Similarities Between the Bucaramanga Earthquake Nest and the Vrancea Seismic Cluster 6.6.4 The Slab Break-off Model 6.6.5 Need of Deep Seismic Refraction and Reflection Data 7. CONCLUSIONS
REFERENCES
227 227 227 231 232 233 233 235 235 235 241 242 242 242 243 243 243 244 244 245 248 248 250 250 250 257 257 257 258 259 261 261
263
xi
ACKNOWLEDGEMENTS The completion of this Ph.D. thesis would have been impossible without the help and cooperation of various people, organizations and institutions. It is a pleasure to acknowledge all people who made possible this thesis, with apologies to anyone I may inadvertently omitted. In the first place I wish to thank my promotor Sierd Cloetingh, for inviting me to do a Ph.D. at the Vrije Universiteit and for helping me in scientific, economical and even personal matters. Also in the first place I wish to thank my copromotor François Roure who encouraged me with the Ph.D. project and offered me his hospitality during three weeks of my work at the IFP. I wish to thank his guidance and his help in many aspects of the project. Jan Diederik van Wees introduced me with the topic of lithospheric stretching and rheological models, he facilitated me the use of the program Whizmod developed by him. He also ran the rheological models discussed in last chapter. Reini Zoetemeijer introduced me with flexural modelling and facilitated me the use of the program Cobra developed by her. Dick Nieuwland introduced me to the analogue modelling topic, made it possible for me to attend to the Thrust Tectonics Conference in London and shared his knowledge and experience on structural geology. Furthermore, I wish to thank all of them for their guidance, stimulating discussions, and constructive criticism on the manuscript. Special thanks for detailed reviews of earlier versions of the manuscript are also due to Harry Doust, Fernando Etayo-Serna, Tomas Villamil, Pedro Restrepo-Pace, and Roberto Linares. I wish to thank my former boss at ICP Kurt Bayer, and Jaime Cadavid Calvo former Director of the Instituto Colombiano del Petróleo (ICP) who approved me a study commission to complete a Ph.D; and Alirio Hernandez Director of the ICP who helped me to get approval for the final expenses. Without the economic support from Ecopetrol and the Ecopetrol-Icetex Fund the completion of this thesis would have been impossible. Egon Castro put special interest in creating possibilities for me to do a Ph.D. and helped me to gather data from Ecopetrol. My former boss at ICP Bernardo Silva, my present boss Yolanda Aguiar, Maria Ximena Mantilla and Alberto Ortiz made possible for me to get additional support for the final corrections, edition and printing of this thesis. Oscar Díaz, Juan Alvaro Gonzalez and Olga Lucia Hernández Pineda helped me with the final edition. Many other friends at Ecopetrol helped me to gather high quality data sets or in many other different ways. Among them Jaime Muñóz, John Ceron, Cesar Mora, Fabio Córdoba, Shajid Kairuz, Ariel Solano, Juan Pablo Reyes, Yolanda Aguiar, Hans Bartels, Myriam Caro, Antonio Rangel, Blanca Nubia Giraldo, Diego García, Felix Texeira, Ivan Olaya and his wife Martha, Oscar Díaz, Alberto Ortíz, Andrés Fajardo, Andrés Reyes, Martín Mantilla, Fernando Munar and many others. Andrés Fajardo supplied me a software for the final correction of the figures. Martín Mantilla supplied me with some seismic lines from the Magdalena Valley. I wish to thank specially to my friends Jorge Rubiano and his wife Myriam for their continuous help in many personal matters in Colombia and for their friendship. Gladys Rocio Ramírez and Bertha Nereida Gómez and many others employees of Ecopetrol, ICP and Icetex made my Ph.D. possible. I acknowledge the Netherlands Research School of Sedimentary Geology and the Vrije Universiteit of Amsterdam for their kind invitation to do a Ph.D. research. I wish to thank in special Anco Lankreijer, for his friendship and kind support, solving me administrative or financial difficulties. I am grateful to all my colleagues in Amsterdam who provided a fruitful and nice working atmosphere and were always ready to help me. Especially my roommates Bernd Andeweg and Ernst Willingshofer solved me many doubts on various matters. Additionally Bernd translated the Dutch summary. I also appreciate the support of my friends Jorge, Eduardo, Sandra and Anton. The support of Fred Beeckman and F. Canemeijer on computer troubles is highly appreciated. Randell Stephenson is thanked for his fruitful comments. Furthermore, I want to thank Aline, Ingrid, Jolante, José, Marlies, Sevgi, Stéphanie, Cees, Daniel, Captain Dick, Gabor, Gerco, Giovani, Harm, Harry, Henk, Joaquim, Markus, Mathias, Christophe, Ritske, Rudie, Taco and Tore for their friendship and help. Margot Saher, Anouk Creusen and Arjan van Vliet helped me with the figures of the last chapter. They were also my roommates during my last months in Amsterdam along with Gideon. I also acknowledge their pleasant company. Thanks also to Alwien Prinsen, Monique Gerbrands, Ellen Salmomé, Liesbeth Aardema, Hetty Turley
xii
and Marjo Duifs who also solved me many practical problems. Sjaak Heezen helped me a lot on logistic matters in Amsterdam before my arrival here. I am indebted with the Institute Français du Petrole (IFP), especially with François Roure who made it possible for me to work there, use their facilities, and participate in a Subtrap meeting at Avignon. Thanks also to William Sassi who helped me using the program Thrustpack and Bernard Colletta for his interest and many suggestions about the structural geology of the Colombian Eastern Cordillera. I like to express my gratitude to Jaime Toro, Nathalie Bordeaux and the Iraqi. All of them made my stay in Paris very pleasant. Thanks also to the employees of the IFP who helped me in financial or administrative matters. I deeply appreciate the unconditional moral support and friendship from José Estevez and his wife Lucy. José Estevez also helped me with some figures. This thesis would have been impossible without the constant support on administrative matters in Colombia from my sister in law Martica Sanchez de Osorio, my friends Jorge Rubiano, his wife Myriam, Hans Bartels and Luis Enrique Cruz. I also deeply acknowledge my mother Lolita, and my sisters and brother. Moral support and love from my wife Gladytas and my sons Cesar and Daniel, is beyond words can describe. I want to dedicate this thesis to them and to the memory of my father.
xiii
SUMMARY The focus of this thesis is on the tectonic basin forming and inversion history of the Eastern Cordillera (EC) of Colombia in terms of the geodynamic processes that govern deformation of the lithosphere. This goal is pursued through compilation of local data into a regional geological model, analysis and quantitative modelling of tectonic subsidence and basin-formation mechanisms. To constrain the alternative possible tectonic scenarios related to the Palaeogene tectonic history, I used quantitative models to test plausible different tectonic scenarios. More than 100 stratigraphic columns and wells of the EC, Llanos Orientales (LLA) and Magdalena Valley (MV) were compiled, as well as previous stratigraphical and paleogeographic interpretations, to draw a series of paleogeographic and original thickness maps, for several time intervals, covering the Mesozoic and Cenozoic. These maps help to unravel the complex rifting and basin inversion history of the EC of Colombia.
Mesozoic rifting history of the Eastern Cordillera. During the Triassic and Jurassic tensional/transtensional stresses, probably initially related to the break-up of Pangea and later to backarc extension, produced lithosphere stretching and generated narrow rifts (< 150 km wide), located in the places of the present day MV and the western flank of the EC (Magdalena-Tablazo sub-basin). During the Early Cretaceous tensional/transtensional stresses probably related to backarc extension produced new episodes of lithosphere stretching and generated a wide (> 180 km wide) system of asymmetric half-rift basins. Subsidence analysis through backstripping and forward modelling of these stratigraphic columns and wells allow to identify five stretching events during the Mesozoic. Using forward modelling techniques, assuming uniform and two layered stretching, I calculated for each stratigraphic column or well the coeval crustal and sub-crustal lithosphere stretching factors. For the Triassic and Jurassic episodes I could only calculate stretching factors for the whole lithosphere due to limited resolution on time data. The main rifted events comprise: (1) Triassic (comprised between ≈248 and 235 Ma, according the geological time scale proposed by Gradstein and Ogg, 1996). This stretching event produced narrow rifts (< 150 km wide), located in the place of the present day Upper Magdalena Valley (with lithosphere stretching factor up to 1.17), the Serranía de San Lucas (lithosphere stretching factor up to 1.23), and less subsiding rift basin at the current location of the western flank of the EC (lithosphere stretching factors up to 1.13). (2) Latest Triassic to Middle Jurassic (comprised between ≈208 and 185 Ma.) and stretching factors up to 1.12 in the Upper Magdalena Valley. (3) Middle Jurassic (comprised between ≈180 and 176 Ma.) with stretching factors up to 1.39 in the western flank of the EC. These two latest Triassic to Middle Jurassic and Middle Jurassic stretching events produced a relatively narrow rift basin (< 150 km wide) at the current location of the north-western flank of the EC, west of Bucaramanga. (Moreover, assuming the thickness of the post-rift section was large along the western flank of the EC north-west of Bogota, part of Jurassic Cretaceous subsidence is likely to relate to the evolution of these early thermal events). In addition, if the thickness of the post-rift Cretaceous section in the western flank of the EC northwest of Bogota is large, it is possible from a two-layered model to hypothesize that this area was thermally weakened by a Jurassic-stretching event. (4) Berriasian to Hauterivian (144 to 127 Ma.). This Early Cretaceous stretching event generated a wide (> 180 km wide) asymmetric half rift basin with a depocenter located along the palaeoeastern flank of the EC and a major normal fault system in its eastern border. A system of horst blocks was also located in the area of the Santander and Floresta massifs. A less developed second order half rift occurred in the place of the southern western flank of the Cordillera. Two layered stretching models with stretching factors up to 1.66 for the crust and up to 3.49 for the subcrustal lithosphere suggest that some decoupling occurred between the crust and subcrustal lithosphere, or that an increased thermal thinning affected the mantle lithosphere. In places of maximum crustal and subcrustal lithosphere stretching (greater than 1.4), small mafic intrusions were emplaced during the Cretaceous (Fabre and Delaloye ,1982).
xiv
(5) Aptian to early Albian (121 to 102.6 Ma.). This stretching event affected mainly the palaeowestern flank of the EC north-west of Bogotá, and the palaeo-Upper Magdalena Valley whereas a major normal fault system was active along the western border of the Cordillera. Subsidence curves suggest that during Late Cretaceous, subsidence was produced by thermal relaxation of the lithosphere. Periods of rift activity correlate in time with gaps of subduction related magmatic arc activity as suggested by Aspden et al. (1987) especially for Jurassic time supporting the hypothesis of back arc extension, which is also supported by volcaniclastic input from the west as indicated by paleocurrent data and local westerly onlap terminations on the basement. If backarc extension continued during the Early Cretaceous by oblique plate convergence, it probably had a strong strike-slip component, as suggested by Aspden et al.. (1987). Initial Triassic rift basins were narrow but increased in width during Triassic and Jurassic times. Cretaceous rifts were wider, and were more asymmetrical than Triassic-Jurassic rift basins. During Berriasian-Hauterivian time the eastern side of the rift possibly was developed by reactivation of an older Palaeozoic rift system associated to the Guaicáramo fault system ( c.f. Hossack, et al., 1999). The western side probably developed by reactivation of an earlier normal fault system developed during Triassic-Jurassic rifting. Probably the increasing width of the rift system was the result of progressive tensional reactivation of pre-existing upper crustal weakness zones. Lateral changes of Mesozoic sediment thickness suggest that the reverse or thrust faults that now define the eastern and western borders of the EC are largelly controlled by former normal faults that became inverted during the Cenozoic Andean orogeny. The oblique orientation of most of them relative to the Mesozoic magmatic arc of the Central Cordillera may be the result of oblique slip extension during Mesozoic, or alternatively can be inherited from the pre-Mesozoic structural grain. However, not all the Mesozoic extensional faults were inverted, some normal faults were passively transported with short-cut basement blocks during Cenozoic inversion ( e.g.E smeraldas Fault, ESRI and Ecopetrol, 1994, Cooper et al., 1995). Repeated Mesozoic stretching events in the same area suggest strain localisation affecting a weak lithosphere. Thermal heating associated with stretching and reactivation of crustal discontinuities probably contributed to such a strain localisation. Thermal processes were more dominant than mechanical stretching during Late Triassic-Early Jurassic phase, in contrast to the Cretaceous rifting phase. During Late Triassic-Early Jurassic abundant volcaniclastic rocks suggest a positive thermal anomaly in the lithosphere but a moderate lithosphere stretching. Triassic-Jurassic age unconformities could have been produced by thermal uplift (“active rifting”?). In contrast during the Cretaceous less abundant volcanic rocks, absence of tectonically controlled unconformities, and the large amount of tectonic subsidence indicates absence of thermal doming. The presence of minor mafic intrusions coinciding with places of maximum crustal and mantle subcrustal stretching, suggest that a modest magmatism took place as a consequence of extension of the lithosphere (“passive rifting”).
Palaeogene incipient basin inversion history of the Eastern Cordillera. During Palaeogene accretion of the oceanic terranes that now form the Western Cordillera correlated with incipient inversion of the Mesozoic extensional basin, where continental to coastal plain and estuarine sediments were deposited. For several time intervals from late Maastrichtian to Early Miocene the tectonic subsidence was calculated and compared to the thermally driven subsidence that would be produced after the Mesozoic stretching events that occurred in the area, assuming a scenario of tectonic quiescence. The great difference between the two scenarii suggests that the hypothesis of tectonic quiescence is not valid for the Palaeogene of the EC. Numerous evidences such as the occurrence of Palaeogene unconformities, the regional Eocene unconformity which locally truncates structures and other local unconformities, lateral changes of facies and thickness, local erosion indicated by detrital composition of sandstone and limited fission track data, all suggest that an incipient inversion of Mesozoic extensional basins occurred during Palaeogene time.
xv
Subsidence in the pre-Andean LLA, palaeo-EC and MV during Palaeogene time cannot be only explained by the topographic load of the palaeo-Central Cordillera as if the basin were a simple foreland basin as interpreted by some authors ( e.g. Cooper et al., 1995). Palaeogene subsidence could only be modelled assuming three different subsidence components: (1) residual thermal subsidence after Mesozoic rifting, (2) flexural subsidence of the lithosphere due to topographic load of the Central Cordillera, and (3) flexural subsidence of the lithosphere produced by incipient topography generated during the Palaeogene in the vicinity of the EC flanks. Flexural subsidence models assuming the lithosphere behaves as an elastic plate of laterally variable thickness, including these three subsidence components along four regional cross sections, suggest that during the Palaeogene local topography (up to 500 m) was developed close to the borders of the former Mesozoic extensional basin probably by inversion along the border extensional faults. However such topography was probably discontinuous and low enough to significantly disturb the sedimentary and palaeocurrent pattern in the Palaeogene basin. Structural kinematic modelling along a regional cross section suggests that the amount of shortening during Palaeogene necessary to produce such Palaeogene topography is small and dependent of the dip angle of the Palaeogene contraction faults. Assuming a dip angle close to 30º and using the structural cross section interpretation by Cooper et al. (1995) the modelled total amount of shortening at the end of Early Miocene was less than 10 km. This early inversion episode accounts for Palaeogene deformation. A direct consequence of this interpretation is the possibility of generation of hydrocarbon traps during Palaeogene, a time when petroleum generation and migration occurred in the place of the EC according to petroleum system modelling results published in the literature ( e.g. Mora, 1996). Palaeogene basin inversion was related to the collision of oceanic plateau terranes with the north-western margin of South America (Nivia, 1987; Kerr et al., 1996, 1997; Sinton et al., 1998). Right-lateral transpressional deformation likely lead to a pre-Andean orogeny in the Central Cordillera during the Palaeogene. Periods of development of basin-inversion and compressional structures seem to correlate with times of high convergence rate. Development of compressional/transpressional structures suggests some mechanical coupling between the orogenic wedge, represented mainly the Central Cordillera, and the regions east of it (MV, EC, and LLA) during Palaeogene time. Modelling results require very low values of effective elastic thickness (EET) in the area of the former extensional basin. The Mesozoic rifting events reduced significantly the strength of the lithosphere, making it very prone to Palaeogene deformation and to further Andean deformation. In the EC area during Palaeogene the reduced EET values (<10 km) inferred from flexural models suggest that lithosphere strength was represented only by the upper crust strength ( c.f. Cloetingh and Burov, 1996). This compressional intraplate deformation was restricted to crustal levels involving “simple shear”-type detachment of the crust at the level of the rheologically weak lower crust from the mantle lithosphere. Incipient inversion of Mesozoic extensional basins occurred by reactivation of previously existing discontinuities, such as Mesozoic extensional faults. The occurrence of earlier deformation in the western side: the MV ( e.g. George et al., 1997; Restrepo-Pace et al., 1999a,b) as compared to the EC or LLA foothills suggests basin inversion “progradation”. This requires that basins located more proximal to the collision front were characterized by a mechanically weaker pre-inversion lithosphere than that of more distal basins.
Neogene basin inversion history of the Eastern Cordillera. During the Neogene the Mesozoic extensional basin was completely inverted. During Middle Miocene to Pliocene Andean deformation and uplift occurred mainly involving inversion of the original Early Cretaceous extensional basin. Uplift of the EC separated the LLA and MV basins. Backstripping analysis of the stratigraphic record of some wells or stratigraphic columns from the EC, LLA and MV and flexural models assuming the lithosphere behaves as an elastic plate has successfully tested the hypothesis that uplift of the EC loaded the lithosphere, thus creating accommodation space in the LLA and MV basins, which were mainly filled with molassic sediments. Flexural subsidence produced by the topographic load of the EC explains the observed Neogene subsidence in the LLA and MV. Irrealistic high bending stresses predicted by flexural models under the EC and the better fit obtained by flexural models assuming a broken plate under the EC support the hypothesis of mantle lithosphere
xvi
subduction (or at least thinning of the lithospheric mantle due to the rise of the isotherms) under the EC. Flexural models results for different stratigraphic markers within the Neogene sedimentary record, and comparison with available fission track data and other evidence of deformation/uplift suggest that during Miocene time important surface uplift (> 1000 m) and exhumation occurred mainly at the locations of the margins of the EC. In the middle axial Tunja-Sabana de Bogotá region where the Palaeogene sedimentary record has been partially preserved, important uplift only occurred later during Pliocene time as recorded by the exceptional palynological record of the Sabana de Bogotá. Such uplifting history can be explained in terms of basin inversion: contractional reactivation of Mesozoic extensional faults initiated during Palaeogene times leading to an initial slight basin inversion, then during Neogene time the compressional deformation increased in rate and magnitude leading to a complete inversion of the original extensional basin. Complete inversion of the master normal fault systems delimiting the former extensional basins probably lead to “extrusion” of the sedimentary fill of half graben basins that now form the eastern and western flanks of the EC. Maximum dip-slip displacement and shortening occurred on the thrust faults connected to inverted Mesozoic normal faults that now approximately delimit the Lower Cretaceous outcrops of both flanks of the EC ( e.g. Colletta et al., 1990; Cooper et al., 1995). In these uplifted flanks Lower Cretaceous or older exposed rocks indicate that all Late Cretaceous or younger sediments have been eroded. Preservation of the Palaeogene sedimentary record in the axial Bogotá-Tunja zone of the EC, as well as the Neogene sedimentary record of the Sabana de Bogotá area suggest that this axial region remained low during Miocene time but was uplifted since the Pliocene. One of the major controls on the effective elastic thickness of the study area is the thermal age of the lithosphere. Based on the thermal age and the values obtained for elastic thickness, three regions can be distinguished in the study area: (1) The LLA basin where effective elastic thickness has a maximum value of 50 to 55 km, the thermal age of its eastern lithosphere being Palaeozoic (this correlation suggests a relatively stabilized strong lithosphere in this area); (2) The EC-MV, where effective elastic thickness has values of 25 km or less, indicating a weak lithosphere, the thermal age of which being about 120 ± 20 Ma (this weak thermal destabilized lithosphere is prone to deformation; weakness there is inherited from the former Mesozoic extensional basins); (3) The Central Cordillera (including westernmost MV) with EET values of 5 km and very young thermal age as indicated by the presence of a recent volcanic arc in this area. EET values lower than 10 km occur in the northern and southern parts of the EC, where major strike-slip faults such as the Santa Marta-Bucaramanga, Bocono and Altamira faults are located. Regionally Andean deformation affected weak lithosphere; however, the weakest lithosphere is associated with strike-slip faults. Such faults probably reach deep enough into the lithosphere to reduce its total strength. Results of flexural modelling indicate crust-mantle decoupling. In the EC mantle began to contribute to lithospheric strength probably only since Neogene time. In the weak EC region, lithospheric strength increased during Neogene time. Two major controls on lithosphere dynamics of the extensional basin formation and inversion history of the EC have been (1) lithosphere rheology, and (2) plate-related tectonic stresses. Standard rheological models and flexural models indicate that during the Mesozoic and Palaeogene a weak lithosphere resulted from lithosphere stretching in the area of Mesozoic extensional basins. However, flexural models indicate a weak lithosphere under the EC in open disagreement with standard rheological models calculated for the present time, that suggest a strong lithosphere in the EC. Two important effects not considered in the standard rheological models may explain the difference: (1) The presence of crustal discontinuities that weaken the lithosphere; and (2) The presence of an anomalous heat input into the lithosphere, as indicated by local Neogene volcanic rocks, and several hot springs in the EC.
Map view restoration on Neogene transpressional basin inversion of the Eastern Cordillera. The Northern Andes of Colombia and Venezuela (where the EC is located) represent a broad zone of deformation resulting from the interaction of the South American, Nazca and Caribbean plates. Such interaction has fragmented the NW corner of the South American Plate into a number of micro plates or tectonic blocks: Guajira, Santa Marta, Maracaibo and Central Colombia (Colombian Central
xvii
Cordillera). Relative movements of the micro-plates resulted in deformed belts with a combination of compressional thick/thin-skinned thrusting, folding and strike-slip faulting which form several mountain ranges or strike-slip fault zones in the Northern Andes. I have manually restored in map view a mosaic of fault-bounded blocks using all the kinematic constraints available in literature, such as strike-slip displacement estimates and amounts of shortening from available structural balanced crosssections. In order to simplify the model I have used a limited number of blocks. This technique is useful to constrain shortening and strike-slip displacement estimates as well as to detect strike-slip motions and rotations about vertical axes not revealed by balancing cross-sections. According to this regional map restoration, the following relative movements of regional blocks leading to deformation belts were interpreted: (1) Eastward convergence and clockwise rotation of the Central Colombia micro-plate relative to South America resulted in transpression of the EC, with northward increasing shortening in this mountain range; (2) Eastward convergence and left-lateral strike-slip movement of Central Colombia relative to Maracaibo produced the northern part of the EC in the Santander Massif region; (3) North-eastward convergence of the Santa Marta Block relative to Maracaibo generated the Perijá Mountain Range; and (4) South-eastward convergence and right-lateral strike-slip motion of the Maracaibo Block relative to South America generated the Mérida Andes of Venezuela. A more detailed map view restoration of the EC of Colombia was done using balanced crosssections from the literature. According to the restoration the amount of shortening during Andean deformation is approximately one half of the present-day width of the Cordillera. Both width and amount of shortening increase northward. Results of the map view restoration are in general supported by kinematic indicators (fault striae) and other outcrop structural data, borehole breakout data and the focal mechanism solutions of upper crustal earthquakes. These results indicate: (1) ENE-WSW shortening perpendicular to the regional structural grain of the EC; (2) Conjugate (?) or pseudoconjugate (?) left-lateral and right-lateral strike-slip faults; (3) Clockwise rotation of the Central Cordillera, MV and western flank of the EC; (4) The SE flank of the EC (Cocuy and eastern Cundinamarca sub-basin) was right-lateral transpressively deformed; (5) The NW flank of the EC (Magdalena-Tablazo sub-basin) was left-lateral transpressively deformed; (6) Andean deformation generated the EC through transpressive inversion of Mesozoic extensional basins. The western part of the EC was affected by N-S sinistral transpression, while the eastern part was affected by E-W dextral transpression. Comparison of displacement of blocks relative to stable South America obtained from the map view restoration, with plate velocity vectors from geophysical data, suggest that the Andes Block as a whole is at present time moving with a dominant right-lateral strike-slip component along the faults at its easternmost boundary. This strike-slip component probably was less important during the whole period of Neogene Andean deformation, as suggested by the map view restoration.
Hypothesis about the deep structure of the Eastern Cordillera. Based on surface geological data, limited geophysical data (gravity, seismicity), comparison with similar mountain belts and analogue and numerical models from the literature, it is possible to narrow the uncertainity range on the deep structure of the EC. Probably it resembles that of the Pyrenees: Lithospheric shortening has been accommodated in the upper brittle crust by development of a double vergent asymmetric wedge, while the mantle lithosphere accommodated shortening by westdipping subduction of the cooler and denser LLA mantle lithosphere under the buoyant and hotter Andean lithosphere. The lower ductile crust probably accommodated shortening by thickening. Probably the EC has been strongly affected by transpression with important transcurrent components. Intermediate seismicity beneath the EC suggest that a subducted slab fragment is present below its NW margin (Taboada et al., 1999, 2000). A small, but very active zone of intermediate-deep seismicity, the Bucaramanga earthquake nest (Schneider et al ., 1987), may result from deep oblique convergence of Palaeo-Caribbean plate fragment connected to the Panamá Block and the Caribbean Plate from the north. This may explain the NW-SE lineation of the nest (Schneider et al ., 1987). Two particularities of the EC, i.e. the local presence of Neogene volcanic rocks at Paipa and Iza, and a intermediate seismicity, may be the expression of slab break-off of an eastward-subducting Caribbeantype mantle lithosphere under the EC. Subduction resistance of the relatively buoyant Caribbean type
xviii
crust (Burke, 1988; Kerr et al ., 1997) may have produced break-off of its denser mantle lithosphere under the Colombian Andes or a fragment of a normal denser lithosphere of the old Farallon Plate. The deep slab suggested by tomographic images (Taboada et al ., 2000) may be a subducted slab, which is no longer connected to the surface plates. This hypothesis would explain the east-dipping slab fragment suggested by seismological studies (Pennington, 1981; Schneider et al ., 1987; Taboada et al ., 1999, 2000) and the Paipa-Iza volcanic rocks. These volcanic rocks may be related with upwelling hot asthenosphere where slab break-off occurs. Possibly partial melting of the lower crust beneath the EC may have generated rising magmas, generated a thermal anomaly under the EC and weakened its lithosphere. Deep seismic studies could offer reliable data in order to support or contradict the suggested hypotheses.
xix
SAMENVATTING Het doel van dit proefschrift is een bijdrage te leveren aan het begrip van hoe verschillende geodynamische processen, die vervorming van de lithosfeer bepalen, hebben bijgedragen aan het ontstaan van een bekken en de verandering van het bekken in een gebergte (inversie) in het gebied van de Cordillera Oriental (Eastern Cordillera, EC) van Colombia. Dit doel is nagestreefd door (1) een compilatie van lokale gegevens in een regionaal geologisch model, (2a) analyse en (2b) kwantitatieve modellering van daling door tektonische processen en bekkenvormende mechanismen. Verschillende mogelijke tektonische scenarios zijn getoetst om te bepalen welke daarvan het meest waarschijnlijk is voor de minder goed bekende tektonische geschiedenis van de EC tijdens het Paleogeen. Meer dan 100 stratigraphische kolommen en putten in de Cordillera Oriental (EC), de Llanos Orientales (LLA) en de Valle del Magdalena (Magdalena Valley, MV) zijn samengevoed met al bestaande stratigrafische en paleogeografische interpretaties om een serie kaarten te construeren voor verschillende tijdsintervallen tijdens het Mesozoï cum en Cenozoï cum van de paleogeografie en oorspronkelijke dikte van de verschillende gesteentepakketten. Deze kaarten helpen om de complexe ontstaans- en vervormingsgeschiedenis van het bekken te ontrafelen.
Mesozoï sche rek geschiedenis van de Cordillera Oriental. Gedurende de Trias en Jura veroorzaakten tensionele en transtensionele spanningen het ontstaan van z.g. "narrow rifts" (<150km breed) in de huidige MV en de Westflank van de EC (Magdalena-Tablazo sub-bekken). Deze spanningen zijn vermoedelijk gerelateerd aan het uiteenvallen van Pangea en vervolgens back-arc extensie. Tijdens het Vroege Krijt veroorzaakten back-arc erelateerde spanningen nieuwe periodes van lithosfeer verdunning in een systeem van z.g. "wide rifts" (>180km breed), met asymmetrische half-grabens. Vijf verschillende Mesozoï sche perioden van rek (‘rifting’) kunnen worden onderscheiden uit een analyse van de dalingsgeschiedenis (‘backstripping’) en door voorwaartse modellering van de stratigrafische kolommen of boringen. Met de voorwaartse modellering, waarbij werd uitgegaan van uniforme rek in twee lagen, kon voor elke stratigrafische kolom of boring bepaald worden met welke factor zowel de korst als het sub-korst gedeelte van de lithosfeer werden verdund (‘stretching factors’). Voor de perioden in Trias en Jura kon door een gebrek aan goede resolutie van de gegevens alleen een algemene stretching factor voor de hele lithosfeer berekend worden. De belangrijkste perioden van rek zijn: (1) Trias (≈248 tot 235 Ma). Deze periode van rek produceerde smalle rek-bekkens (< 150 km breed), gelokaliseerd op de huidige plaats van de MV (met lithosfeer stretching factor tot 1.17), de Serranía de San Lucas (lithosfeer stretching factor tot 1.23), en een minder hard dalend rek-bekken op de huidige locatie van de westelijke flank van de EC (lithosfeer stretching factors tot 1.13); (2) Laatste Trias tot Midden Jura ( ≈208 tot 185 Ma.) met stretching factors tot 1.12 in de Upper MV; (3) Midden Jura (≈180 tot 176 Ma.) met stretching factors tot 1.39 in de westelijke flank van de EC. Deze twee perioden van rek leidden tot een relatief smal bekken (< 150 km wiid) op de positie van de huidige noordwestelijke flank van de EC, ten westen van Bucaramanga. (Als aangenomen wordt dat de dikte van de post-rift section aanzienlijk was langs de westelijke flank van de EC in dit gebied ten noord-westen van Bogota, kan met een model dat uitgaat van twee sterke lagen in de lithosfeer, de hypothese worden opgesteld dat dit gebied thermisch verzwakt werd door een riftperiode in de Jura. Een deel van de daling in Jura en Krijt is waarschijnlijk gerelateerd aan de ontwikkeling van deze vroege thermische events). (4) Berriasian tot Hauterivian (144 tot 127 Ma.). Deze Vroeg Krijt riftperiode genereerde een breed (> 180 km breed) asymmetrisch half rek-bekken met een depocenter langs de paleo-oostelijke flank van de EC en een groot systeem van afschuivingsbreuken langs de oostelijke rand. Andere gebieden waarin een systeem van horsten ontstond zijn de tegenwoordige massieven van Santander en Floresta. Een model van rek in twee lagen leverde stretching factors op van 1.66 voor de korst en tot 3.49 voor het subkorst deel van de lithosfeer. Dit suggereert dat een zekere mate van ontkoppeling plaats vond tussen de korst en het lithosfeer gedeelte onder de korst, of dat xx
een verhoogde thermische activiteit het mantel gedeelte van de lithosfeer verdund heeft. Op de plekken met maximale stretchings factors ontstonden kleine mafische intrusies tijdens het Krijt (Fabre and Delaloye ,1982); (5) Aptian tot Vroeg Albian (121 tot 102.6 Ma.). Deze periode van rek beï nvloedde voornamelijk de paleo-westelijke flank van de EC ten noord-westen van Bogotá, en de paleo-Valle Superior del Magdalena, terwijl een belangrijk systeem van afschuivingsbreuken actief was langs de westelijke grens van de EC. Dalings curves suggereren dat tijdens Laat Krijt daling werd veroorzaakt door thermische relaxatie van de lithosfeer. Perioden in rift-activiteit correleren met tijdsintervallen waarin de activiteit van de vulkaanboog afwezig was volgens Aspden et al . (1987), in het bijzonder tijdens de Jura. Dit ondersteunt de hypothese dat de rift ontstond als een ‘back-arc’ extensie, rek achter een subductie gerelateerde boog, wat ook blijkt uit input van volcaniclastic uit het westen (zoals aangetoond door gegevens over paleostroomtrichting) en lokaal westwaarts ‘onlap terminations’ over het basement. Als back-arc extensie gedurende het Vroeg Krijt doorging door zijdelingse convergentie van platen, dan had het daardoor een grote zijwaartse component, zoals gesuggereerd door Aspden et al . (1987). De rek-bekkens waren oorspronkelijk in het Trias smal, maar verbreedden zich gedurende Trias-Jura. De Krijt bekkens waren nog breder en meer asymmetrisch dan de Trias-Jura rek-bekkens. Tijdens Berriasian-Hauterivian ontwikkelde de oostzijde van het bekken zich waarschijnlijk door reactivatie van een Paleozoisch rift systeem, geassocieerd met het Guaicáramo breuk systeem ( c.f. Hossack, et al., 1999). Ook de westelijke zijde ontwikkelde zich waarschijnlijk door reactivatie, in dit geval reactivatie van een systeem afschuivingsbreuken dat was ontstaan tijdens de rek in Trias en Jura. De verbreding van het systeem was een resultaat van progressieve reactivatie onder rek van reeds bestaande zwaktezones in de bovenkorst. Laterale veranderingen in de dikte van de Mesozoï sche sedimenten wijzen er op dat de over- en opschuivingsbreuken die nu de oostelijke en westelijke rand vormen van de EC voornamelijk zijn ontstaan uit oudere afschuivingen die werden geï nverteerd tijdens de Andean gebergte vorming in het Cenozoï cum. Het hoekverschil tussen de meeste breuken en de a) zou het resultaat kunnen zijn van ‘oblique slip’ rek tijdens het Mesozoï cum, of als alternatief, overgenomen van al bestaande structuren in de korst voor het Mesozoï cum. Niet alle Mesozoï sche afschuivingbreuken werden overigens gereactiveerd, sommigen werden passief meegevoerd in basement blokken tijdens de inversie in het Cenozoï cum. Herhaalde perioden van rek in het Mesozoï cum in dezelfde gebieden suggereren dat de vervorming gelokaliseerd werd in zwakke delen van de lithosfeer. Opwarming die samenhangt met rek en reactivatie van discontinuï teiten in de korst hebben waarschijnlijk bijgedragen aan deze lokalisatie van vervorming. Het veelvuldig voorkomen van Laat Trias tot Vroeg Jura volcaniclastische gesteenten suggereren een positieve thermische anomalie in de lithosfeer, maar slechts beperkte lithosferische rek. Unconformiteiten van Trias-Jura ouderdom zouden gevormd kunnen zijn door thermische opheffing (active rifting?). Thermische processen waren dominanter dan mechanische rek tijdens Laat Trias tot Vroeg Jura, in tegenstelling tot de rekfase in het Krijt toen thermische opheffing waarschijnlijk afwezig was, zoals blijkt uit de volgende feiten: minder veel voorkomen van volcaniclastische gesteenten, de afwezigheid van unconformiteiten -veroorzaakt door tektoniek- en de grote tektonische daling. De aanwezigheid van kleine mafische intrusies komen overeen met regio’s van maximale korst en mantel rek, wat er op wijst dat beperkte magmatische activiteit plaats vond als gevolg van rek in de lithosfeer (passieve rifting).
Paleogene start van bekken-inversie in de EC. Tijdens het Paleogeen correleert de accretie van fragmenten oceanische korst, die nu de Westelijke Cordillera vormt, met het begin van de inversie van het Mesozoï sche rek-bekken waarin continentale tot kustvlakte en estuariene sedimenten werden afgezet. Voor verschillende tijdsintervallen van Laat Maastrichtian tot Vroeg Mioceen is de tektonische daling berekend en vergeleken met de daling gedreven door thermische activiteit die zou worden geproduceerd na de rekperiode die plaatsvond in het gebied, uitgaand van een scenario met tektonische xxi
rust na de rek. De grote verschillen tussen de twee scenario’s suggereren dat de hypothese van tektonische rust niet geldig kan zijn voor het Paleogeen in de EC. Verschillende feiten wijzen ook op een beginnende inversie van het Mesozoï sche rek-bekken tijdens het Paleogeen: (1) Paleogene unconformiteiten, (2) de regionale Eocene unconformiteit die lokaal oude structuren en unconformiteiten afsnijdt, (3) laterale verandering van facies en dikte, (4) lokale erosie als aangetoond door de detritische samenstelling van zandsteen, en (5) beperkte gegevens van een splijtsporen-analyse. Daling in de pre-Andean LLA, EC en MV tijdens het Paleogeen kan niet alleen worden verklaard door topografische lading van de Centrale Cordillera (CC), alsof het bekken een eenvoudig voorlandbekken zou zijn, zoals wel geï nterpreteerd is door sommige auteurs (e.g . Cooper et al.,1 995). Paleogene daling kan alleen gemodelleerd worden door drie verschillende componenten van de daling mee te nemen: (1) een residu van thermische daling na de Mesozoï sche rek (2) daling ten gevolge van flexuur van de lithosfeer door topografische lading door de CC en (3) daling ten gevolge van topografie die ontstond tijdens het Paleogeen in de buurt van de flanken van de EC. Dalingsmodellen die flexuur gebruiken nemen aan dat de lithosfeer zich gedraagt als een elastische plaat van een eventueel lateraal variabele dikte. Als we de drie genoemde componenten van daling mee nemen in vier regionale profielen, suggereert dat dat de Paleogene locale topografie (tot 500m hoog) ontstond dicht bij de randen van het vroegere Mesozoï sche rek-bekken, waarschijnlijk door inversie van de randbreuken. De topografie was waarschijnlijk discontinu en laag genoeg om het sedimentaire patroon en paleostroomrichting van het Paleogene bekken behoorlijk te beï nvloeden. Kinematische modellering langs een regionale sectie suggereren dat er maar weinig verkorting nodig was om deze Paleogene topografie te vormen, afhankelijk van de hellingshoek van de breuken waarlangs de verkorting plaatsvindt. Uitgaande van een hellingshoek dicht bij 30 graden, en gebruikmakend van de interpretatie van een structureel profiel door Cooper et al (1995), is de gemodelleerde hoeveelheid verkorting aan het eind van het Vroeg Mioceen minder dan 10 km. Een direct gevolg van deze interpretatie is de mogelijkheid van ontwikkeling van ‘hydrocarbon traps’ tijdens het Paleogeen, een tijdsinterval waarin volgens de literatuur ontwikkeling en migratie van petroleum plaatsvond in de EC ( e.g . Mora, 1996). Paleogene bekken-inversie was gerelateerd aan de botsing van oceanische plaatfragmenten met de noordwestelijke marge van Zuid Amerika (Nivia, 1987; Kerr et al., 1996, 1997; Sinton et al., 1998). Rechtsom zijwaartse transpressieve vervorming leidde vermoedelijk tot een pre-Andean gebergtevorming in de CC tijdens het Paleogeen. Perioden van ontwikkeling van bekken-inversie en compressieve structuren lijken overeen te komen met perioden met een hoge convergentie-snelheid. De ontwikkeling van compressieve of transpressieve structuren suggereert een zekere mechanische koppeling tussen de orogene ‘wig’, bestaand uit voornamelijk de CC, en de regio’s oostelijk van de CC tijdens het Paleogeen (MV, EC en LLA). De modelresultaten vereisten erg lage waarden voor de Effectieve Elastische Dikte (EET) in het gebied van het voormalige rek-bekken. De Mesozoï sche rekperioden reduceerden de sterkte van de lithosfeer aanzienlijk, waardoor het gebied zowel in het Paleogeen als de Andean gebergtevorming betrokken raakte. Tijdens het Paleogeen werd de sterkte van de lithosfeer bepaald door alleen de sterkte van de bovenkorst ( c.f. Cloetingh and Burov, 1996), volgens de waarden voor de EET die voortkomen uit de flexuurmodellen. Deze compressieve intraplate vervorming was beperkt tot de korst niveaus door middel van ‘simple shear’ type ontkoppeling van de korst en mantel lithosfeer langs het niveau van de zwakke onderkorst. Het eerder voorkomen van vervorming in de westelijke zijde (MV, e.g. George et al., 1997; Restrepo-Pace et al., 1999a,b) in vergelijking met de EC of LLA suggereren een progradatie van bekken-inversie. Dit vereist dat de bekkens die dichter bij de botsing waren gelokaliseerd, gekarakteriseerd worden door een mechanisch zwakkere lithosfeer voor de inversie dan de verder weg gelegen bekkens.
Neogene bekken-inversie in de Cordillera Oriental. Tijdens het Neogeen werd het Mesozoï sche rek-bekken volledig geï nverteerd. Andean vervorming en opheffing gebeurde van Midden Mioceen tot Plioceen vooral door de inversie van het originele Vroeg Krijt rek-bekken. Opheffing van de EC scheidde het LLA- van het MV-bekken. Zowel backstripping van de stratigrafische opeenvolging van een aantal putten en stratigrafische kolommen van de EC, LLA en MV als flexuurmodellen laten zien dat de opheffing van de EC de lithosfeer laadde xxii
zodat er accomodatie ruimte ontstond in de LLA en MV bekkens die voornamelijk werden gevuld met molasse sedimenten. Onrealistisch hoge buigspanningen die door flexuurmodellen voorspeld worden onder de EC en de betere overeenstemming met de waarnemingen behaald met flexuurmodellen die uitgaan van een gebroken plaat onder de EC, steunen de hypothese van mantel lithosfeer subductie (of op zijn minst verdunning van het mantel gedeelte van de lithosfeer door een stijging van de isothermen). De resultaten van de flexuurmodellen voor de verschillende stratigrafische niveaus in de Neogene sedimentaire opeenvolging en vergelijking met beschikbare splijtingssporen gegevens en andere aanwijzingen voor deformatie en/of opheffing suggereren dat tijdens Mioceen belangrijke opheffing en exhumatie (tot >1000m) plaatsvond, voornamelijk langs de marges van de EC. In de axiale regio (Tunja-Sabana de Bogotá), waar de Paleogene opeenvolging gedeeltelijk bewaard is gebleven vond belangrijke opheffing pas plaats tijdens het Plioceen, zoals is achterhaald uit de buitengewone palynologische gegevens van de Sabana de Bogotá. Een dergelijke geschiedenis van opheffing kan worden uitgelegd in termen van bekken inversie: reactivatie onder verkorting van de Mesozoï sche rekbreuken begon tijdens het Paleogeen en leidde tot een initiëel slecht beperkte inversie van het bekken, met daarna tijdens het Neogeen een volledige inversie van het bekken. Volledige inversie van de grote systemen van afschuivingsbreuken die de randen vormden van het vroegere bekken, leidde waarschijnlijk tot extrusie van de sedimentaire vulling van de half-graben bekkens die nu de oostelijke en westelijke flanken van de EC vormen. Een maximum verplaatsing langs de helling van de breuk en verkorting vond plaats langs de opschuivingen die behoorde bij de geï nverteerde Mesozoï sche afschuivingen die nu ongeveer de voorkomens van Onder Krijt ontsluitingen in beide flanken van de EC begrenzen ( e.g. Colletta et al., 1990; Cooper et al., 1995). In deze opgeheven flanken zijn Onder Krijt en oudere gesteente zichtbaar die aangeven dat alle Laat Krijt en jongere sedimenten zijn geërodeerd. Dat de Paleogene opeenvolging is bewaard gebleven in de axiale Bogotá-Tunja zone van de EC, als ook de Neogene sedimenten in de Sabana de Bogotá, wijst erop dat dit gebied laag gebleven is tijdens het Mioceen en pas werd opgeheven vanaf het Plioceen. Een van de belangrijkste factoren die de EET bepaalt in het bestudeerde gebied is de thermische ouderdom van de lithosfeer. Gebaseerd op de thermische ouderdom en de waarden verkregen voor de EET, kunnen drie regio’s onderscheiden worden: (1) het LLA bekken met EET waardes tot 50-55km en een thermische ouderdom van Paleozoicum. Dit wijst samen op een relatief stabiele en sterke lithosfeer, (2) EC en MV, waar de EET ongeveer 25km bedraagt en de thermische ouderdom om en nabij 120 Ma is. Dit duidt op een thermische gedestabiliseerde lithosfeer die gevoelig is voor vervorming; en (3) de CC (inclusief het meest westelijke deel van MV) met EET waardes van slechts 5 km en erg lage thermisch ouderdom, zoals blijkt uit de aanwezigheid van een recente vulkanische boog in het gebied. Waarden van minder dan 10km voor de EET komen voor in de noordelijke en zuidelijke gedeelten van de EC, op plekken waar grote zijschuivingen zoals de Santa Marta-Bucaramanga, Bocono en Altamira breuken te vinden zijn. Zulke breuken reiken waarschijnlijk diep genoeg in de lithosfeer om die te kunnen verzwakken. Resultaten van de flexuur modellering wijzen op ontkoppeling tussen korst en mantel. In de EC nam de sterkte van de lithosfeer toe in het Neogeen, meest waarschijnlijk begon de mantel toen pas mee te tellen in de sterkte van de hele lithosfeer. Twee grote factoren die invloed hebben op de lithosfeer dynamiek tijdens de vorming van rek bekkens en de inversie van de EC zijn: (1) de rheologie (materiaal eigenschappen) van de lithosfeer, en (2) tektonische spanningen, samenhangend met plaat-interactie. Standaard rheologie- en flexuurmodellen geven aan dat een zwakke lithosfeer tijdens Mesozoï cum en Paleogeen het resultaat was van de Mesozoï sche rek. De flexuurmodellen wijzen echter op een zwakke lithosfeer onder deEC voor de huidige situatie, wat in tegenspraak is met de uitkomsten van standaard rheologische modellen voor dezelfde situatie: die suggereren een sterke lithosfeer. Twee belangrijke effecten die niet meegenomen zijn in de standaard rheologische modellen zouden de verschillen kunnen verklaren: (1) de aanwezigheid van discontinuï teiten in de korst die de lithosfeer verzwakken, en (2) de aanwezigheid van een anomale hittebron, zoals aangetoond door Neogene vulkanische gesteenten en verschillende hete bronnen in de EC.
xxiii
Reconstructie in kaartbeelden van de Neogene transpressieve inversie van het bekken van de EC. De noordelijke Andes van Colombia en Venezuela (waar de EC zich bevindt), vormt een brede zone van vervorming als gevolg van interactie tussen de Zuid-Amerikaanse, Nazca en Caribbische platen. Deze interactie heeft ervoor gezorgd dat de NW hoek van de Zuid-Amerikaanse plaat is opgebroken in verschillende fragmenten: Guajira, Santa Marta, Maracaibo en Centraal Colombia (Centraal Colombiaanse Cordillera). Relatieve bewegingen van deze micro-platen zorgde in de noordelijke Andes voor gebergteketens en zones met een combinatie van compressieve thick/thinskinned thrusting, plooiï ng en zijschuivingsbreuken. Om de complexiteit van het model binnen de perken te houden, is maar een beperkt aantal blokken die door de breuken van elkaar gescheiden worden gebruikt in een handmatige reconstructie. Deze reconstructie is gemaakt aan de hand van kinematische randvoorwaarden die in literatuur gevonden konden worden, zoals de verplaatsing van de zijschuivingen en schattingen van verplaasting uit gereconstrueerde profielen door het gebied. Deze techniek is zeer nuttig om betere schattingen te kunnen maken van de verkorting en verplaatsing en om rotaties te ontekken die niet in profielen kunnen worden vastgesteld. Volgens deze regionale reconstructie kunnen de volgende bewegingen van de blokken die hebben bijgedragen aan de vormingen van de gebergte- gordels, worden vastgesteld: (1) oostwaartse convergentie en rotatie met de klok mee van de Centraal Colombiaanse micro-plaat ten opzichte van Zuid-Amerika, wat resulteerde in transpressie in de EC met naar het noorden toenemende verkorting; (2) oostwaartse convergentie en linksom zijwaartse beweging van Centraal Colombia ten opzichte van Maracaibo, wat het noordelijk deel van de EC in de regio van het Santander massief vormde; (3) noord-oostelijke convergentie van het Santa Marta Blok ten opzichte van Maracaibo zorgde voor het ontstaan van het Perijá gebergte; en (4) zuid-oostwaartse convergentie en rechtsom zijwaartse beweging van het Maracaibo Blok ten opzichte van Zuid-Amerika. Dit veroorzaakte de Mérida Andes van Venezuela. Een meer gedetailleerde reconstructie op kaartbeelden is uitgevoerd voor de EC, gebruikmakend van profielen uit de literatuur. Volgens deze reconstructie is de totale verkorting tijdens de Andes vervorming ongeveer gelijk aan de helft van de huidige breedte van de Cordillera. Zowel breedte als verkorting nemen toe richting het noorden. Deze resultaten worden over het algemeen ondersteund door bewegings-indicatoren (bewgingsrichting op breukvlakken) en andere stuctureel geologische gegevens uit ontsluiting, vervorming van boorgaten, en gegevens over de verkortingsrichting verkregen uit aardbevingen in de bovenkorst. De reconstructie toont dat: (1) ONOWZW verkorting plaatsvond, loodrecht op de regionale structuur van de EC; (2) conjugate (?) of pseudo-conjugate (?) linksom en rechtsom zijschuivingen zich ontwikkelden; (3) de CC, MV en westelijke flank van de EC met de klok meegedraaid zijn; (4) de zuidoostelijke flank van de EC (Cocuy en oostelijke Cundinamarca sub-bekken) vervormd werd door transpressie met een grote component rechtsom; (5) de noordwestelijke flank van de EC (Magdalena-Tablazo sub-bekken) juist vervormd werd door transpressie met een grote component linksom; (6) vervorming van de Andes zorgde voor transpressieve inversie van de oorspronkelijke Mesozoï sche rek-bekkens. Het westelijke deel van de EC werd beï nvloed door noord-zuid georienteerde transpressie met linksom beweging, terwijl het oostelijk deel oost-west georienteerde transpressie met een component rechtsom onderging. Vergelijking van waarden voor de verplaatsing van blokken ten opzichte van het stabiele ZuidAmerika die zijn verkregen door de kaartrecontructies met plaatsnelheden van geofysische gegevens, suggereren dat het Andes Blok als geheel op het moment met een dominante rechtsom zijwaartse beweging langs de oostelijke delen beweegt. Deze component van zijwaartse beweging was waarschijnlijk minder belangrijk tijdens de gehele periode van Neogene vervorming van de Andes, volgens de reconstructie op kaartbeeld.
Hypothese over de diepe structuur van de EC. Gebaseerd op geologische gegevens, beperkte geofysische date (zwaartekracht en seismiciteit), vergelijking met vergelijkbare gebergten en analoge en numerieke modellen uit de literatuur, is het mogelijk de mogelijke variatie in diepe structuur van de EC. Waarschijnlijk lijkt de diepe structuur van de EC nogal op die van de Pyreneëen. Verkorting van de lithosfeer heeft plaatsgevonden in de brosse boven korst door een dubbel-vergente asymmetrische wig, terwijl op het niveau van de mantel xxiv
verkorting mogelijk is geworden door westwaartse subdcutie van de koudere en zwaardere mantel lithosfeer van de LLA onder de bouyant en warmere Andean litosfeer. De ductiele onderkorst heeft de verkorting waarschijnlijk opgevangen door verdikking. Waarschijnlijk is de EC sterk beinvloed door transpressie met een belangrijke transcurrent component. Seismiciteit van gemiddelde diepte onder de EC suggereert dat een gesubduceerd deel van de lithosfeer nog aanwezig is onder de NW rand (Taboada et al., 1999, 2000). Een kleine, maar actieve zone met aardbevingen van gemiddelde diepte (de Bucaramanga earthquake nest (Schneider et al., 1987), kan het resultaat zijn van scheve convergentie van de oude Caribische plaat, verbonden aan de Panama en de recente Caribische plaat. Dit kan de NW-ZO verspreiding van deze aardbevingen verklaren, (Schneider et al., 1987). Twee opvallende eigenschappen van de EC, n.l. de aanwezigheid van Neogene vulkanische gesteenten (in Paipa en Iza ) en de seismiciteit van gemiddelde diepte, kunnen een aanwijzing vormen voor het afbreken van het mantel gedeelte van de oostwaards gesubduceerde Caribische plaat onder de EC. Subductie weerstand van de relatief lichte Caribische korst (Burke, 1988; Kerr et al., 1997) kan hebben geleid tot het afbreken van de dichtere lithosfeer van de oude Farallon plaat. De diepe slab, gesuggereerd op basis van tomgrafische beelden (Taboada et al., 2000) kan een gesubduceerde plaat zijn, die niet langer is gekoppeld aan oppervlakte plaaten. Deze hypothese kan de oostduikende plaat uit seismologische studies van (Pennington, 1981; Schneider et al., 1987; Taboada et al., 1999, 2000) en de Paipa-Iza vulkanische gesteenten verklaren. De vulkanische gesteenten kunnen gerelateerd zijn aan de stijgende hete asthenosfeer op plaatsen waar de plaat gebroken is. Mogelijke gedeeltelijke smelt van de onderkorst onder de EC kan hebben geleid tot magma vorming, thermische anomalieën onder de EC en plaatselijke verzwakking van de lithosfeer. Diep seismische studies zouden betrouwbare gegevens kunnen opleveren om deze hypotheses te falsifiëren of verifiëren.
xxv
CHAPTER 1 INTRODUCTION AND OUTLINE 1.1. AIM AND METHODS The main aim of this thesis is to contribute to understand the tectonic basin forming and inversion history of the Eastern Cordillera (EC) of Colombia in terms of the geodynamic processes that govern deformation of the lithosphere. This goal is pursued through compilation of local data into a regional geological model, analysis and quantitative modelling of tectonic subsidence and basinformation mechanisms. To constrain the alternative possible tectonic scenarios related to the not-well constrained Palaeogene tectonic history, I used quantitative models to test plausible different tectonic scenarios. Details of the methods are explained in the chapter in which they are applied.
1.2 BASIN FORMATION AND INVERSION MODELS Sedimentary basin formation mechanisms are basically: (1) Thermal mechanisms producing subsidence or uplift due to their effects on the lithosphere density structure (e.g.. Sleep, 1971) and its isostatic balance. An example of this is the cooling and subsidence of oceanic lithosphere as it moves away from spreading centers. (2) Changes in lithosphere thickness and structure produced by stretching, erosion or emplacement of dense material into the lithosphere column leading to isostatic readjustment. An example of this is mechanical stretching of the crust accompanied by normal faulting occurring in extensional basins (e.g.. McKenzie, 1978). (3) Loading of the lithosphere causing lateral subsidence due to regional isostatic effects (flexure) by mountain building processes, as occurring in foreland basins ( c.f.P rice, 1973). In many cases where the lithosphere is affected by rapid stretching, as the case of rifting, lithospheric thinning and upwelling of hot asthenosphere change its density and thermal structure. The consequent isostatic readjustment causes fast tectonic subsidence followed by slower thermal subsidence as the lithosphere thickness gradually increases by heat conduction (McKenzie, 1978). The simple stretching model of McKenzie (1978) has been followed by more sophisticated quantitative models, including differential stretching finite rates of extension (Jarvis and McKenzie, 1978), lateral heat conduction (Royden and Keen, 1980), depth dependent rheology (Sawyer, 1985), and other complex effects. Assumption of local isostasy implies lack of lateral strength of the lithosphere. More realistic models assume the lithosphere behaves to applied loads as an elastic plate overlying a weak asthenospheric mantle. This has led to development of the concept of flexural subsidence that can explain subsidence observed in foreland basins (Beaumont, 1981). Initial flexural models applied to foreland basins assumed only topographic loads (Beaumont, 1981). However, later models have suggested other possible subsurface loads. The strength of the lithosphere controls its flexural behaviour. The distribution of strength in the lithosphere varies vertically, due to its petrological and thermal structure (Ranalli and Murphy, 1987; Ranalli, 1995; Cloetingh and Burov, 1996), and laterally as a result of its pre-orogenic history (De Jong, 1991; Van Wees et al ., 1992; Van Wees and Cloetingh, 1993; Van Wees, 1994; Ziegler et al ., 1995, 1998). Extrapolation of rock mechanical, geophysical (including deep seismic) and geological data support the concept of a rheologically stratified lithosphere characterized by a relatively strong (mostly brittle) upper crust, a weak (mostly ductile) lower crust, and a strong (brittle and ductile) subcrustal lithosphere (Ranalli, 1995). This will be discussed in last chapter. During basin inversion, basin controlling faults reverse their movement due to compressional tectonics, and basins become positive topographic features ( e.g.. Bally, 1984). Numerical models for thermal compression and exhumation of the lithosphere (England and Thompson, 1986; Davy and Gillet 1986) have been applied to compressional settings such as those of mountain building and basin inversion. However, models for the rheological evolution of the lithosphere indicate that the predicted
Introduction Outline
rheological evolution cannot explain repeated reactivation of faults (Van Wees, 1994; Ziegler et al., 1995, 1998). Observed reactivation of upper crustal faults not aligned with respect to the stress field, deformation patterns in rifted and inverted basins and other evidence, suggest that in addition to the standard rheology of the lithosphere, also reactivation of inherited weak fault zones control the response of the lithosphere to tectonic stress fields (Van Wees, 1994; Ziegler et al., 1995, 1998). To a large extent, the mechanical processes responsible for sedimentary basin formation and evolution are thermo-mechanical, inducing deformation in the lithosphere, and related to the presence of pre-existing discontinuities that have a crucial role in weakening the lithosphere (Van Wees, 1994; Ziegler et al., 1995, 1998). In a sedimentary basin the stratigraphic record, including sedimentary fill, unconformities and other indicators of uplift, reflects the history of vertical movements. The study of the sedimentary fill and its associated tectonic subsidence signal gives important information about basin formation mechanisms (Ter Voorde, 1996; references in Cloetingh et al., 1993). Several palaeo-thermal indicators, such as fission tracks, vitrinite reflectance and mineral palaeo-thermometers are available to record thermal history.
1.3 THE EASTERN CORDILLERA OF COLOMBIA The Eastern Cordillera (EC) of Colombia in an excellent example of a Mesozoic extensional basin inverted during Cenozoic that can be used as a natural laboratory to study these processes. In the EC the stratigraphic record is better constrained than other uplift or thermal indicators. For this reason in the EC and neighbouring areas the stratigraphic record constitutes the natural starting point to study vertical movements induced by tectonic processes responsible for basin formation, with the aim to infer from them the thermo-mechanical behaviour of the lithosphere. Although several conceptual tectonic models have been proposed for the evolution of the EC (Julivert, 1970; Fabre, 1987; Colletta et al., 1990; Dengo and Covey, 1993; Cooper et al., 1995; Roeder and Chamberlain, 1995; Casero et al., 1997; Branquet, 1999), the tectonic basin forming and inversion history of the EC of Colombia has not been studied in terms of the geodynamic processes that govern deformation of the lithosphere. The deep structure of the EC is largely unknown. Although there is some geological evidence of basin inversion during the Palaeogene, such inversion is poorly documented and not regionally understood. Additionally important basic data are not available for the EC: For example there are no thickness maps for the Meso-Cenozoic sedimentary fill in the area.
1.4 TECTONIC SETTING OF THE EASTERN CORDILLERA The EC is the eastern branch of the Colombian Ande s (Fig. 1.1). The latter comprise three mountain ranges: the Eastern, Central and Western Cordilleras , which merge southward into a single range. The Magdalena Valley (MV) separates the Eastern and Central Cordilleras and the Cauca Valley separates the Central and Western Cordilleras. The EC and its bounding basins, the Llanos Orientales (LLA) in the east and MV in the west define the area studied in this thesis. During the Mesozoic the area of the EC was an extensional basin. During the Palaeogene some authors (e.g.. Van der Hammen, 1961; Roeder and Chamberlain, 1965; Restrepo-Pace et al ., 1999a,b; Gomez et al ., 1999) have suggested the possibility of upthrusted blocks and/or incipient inversion of the Mesozoic extensional basin in the area of the EC. However a commonly accepted view is that of a unique simple foreland basin related to the top ographic load of the Central Cordillera ( e.g. Cooper et al ., 1995). This problem will be discussed in Chapter 3. During the Neogene there is a general agreement that the Mesozoic extensional basin became inverted, deformed and uplifted to form the EC (Cooper et al., 1995). In the study area, during Triassic and Jurassic, continental and volcanic facies were deposited in narrow rifts developed behind a magmatic arc related to the subduction of the Pacific plates under the western border of South America. During Early Cretaceous marine facies were accumulated on a wide rift system, probably a back-arc basin related to subduction of Pacific plates west of Colombia (Aspden et al., 1987). Shallow marine sedimentation ended at the end of the Cretaceous due to the
2
Chapter 1
78
0
0
76
0
e a S a n e b r i b a C
12
y a l l e V t l n a e B e d a l t o n i c a g M t J a l e B a n e r w S u i n L o
0
10
P a n a m a 8
74
0
S S e r r a n i a
y d le e
6
4
2
0
0
0
n a e c O c i f f i c a P i n a s B i c f i f a c P
B a u c o
l a V
to t ra A
n a u J n a S
n e r t e s W
r a l le i rd o C
r a l e l i r d o C
0
0
72 0
680
70
r a a j i u a i r G a j G u a s i n B
a a r t r r d a a i e v a M S e t a N n a S d e
j a y i r e l e l P a V e d a r a i s e a n C r r e S
a n l e ) a d V g M a ( M y e l e l d V a i d M
n e r s t a E
r a l e i l d r C o
M
a i d e r M
s d e A n
Venezuela Barinas Basin
) E C (
Llanos Orientales Basin (LLA)
y l e l a V a u c l a C r a
y l e l a V
a n n e t l a e n d C a g M
L i n e o f S e c t i o n F i g . 1 . 2
Study area (Fig. 2.1)
r e p p U
Guyana
Shield
Putumayo Am a zo na s
Basin
0
0
Basin
Ecuador
0
100
200 Km
Figure 1.1: Location Map accretion of the oceanic terranes of the Western Cordillera ( Fig. 1.2, Cooper et al., 1995). According to Cooper et al., (1995), accretion of the Western Cordillera created the early foreland basin during the Palaeogene (Fig. 1.2). A different scenario for the Palaeogene is one of incipient inversion of the Mesozoic extensional basin (Restrepo-Pace et al ., 1999a,b; Gomez et al ., 1999). During the Neogene (since middle Miocene) the onset of Andean deformation in the EC isolated the MV basin from the LLA basin ( Fig. 1.2). The deformation was dominated by inversion of the basin controlling faults to create the EC (Cooper et al., 1995; Fajardo-Peña, 1998).
1.5 THESIS OUTLINE This thesis gives an overview of the stratigraphy and structural geology of the EC and summarises from literature its geological evolution. In order to use the stratigraphic record of the EC and its neighbouring areas to study tectonic vertical movements and basin forming mechanisms, I
3
Introduction Outline
compiled from the available literature more than 100 stratigraphic columns and wells. I also used previous stratigraphic and palaeogeographic interpretations to create a series of palaeogeographic and restor ed original thickness maps for several time intervals, covering the Mesozoic and Cenozoic. Chapter 2 su mmarises the Mesozoic stratigraphy and discusses the vertical (tectonic subsidence) and horizontal (lithosphere stretching) movements that occurred in the Colombian basin during the Mesozoic. At that time the Colombian basin was an extensional basin. The reconstruction is accomplished by using the documented sedimentary record as an indicator of subsidence. Because basin subsidence is produced both by tectonic processes responsible for basin formation, as well as by isostatic compensation of surface loads produced by sediment and water filling the basin, it is necessary to separate the tectonic signal from the total subsidence signal. I applied 1D-backstripping techniques (Bond and Kominz, 1984) in order to do this. Theoretically, if the eustatic variation through time is known, it is possible to quantify the contribution to subsidence produced by water loads. However, quantification of the sea level curve is difficult and debatable ( e.g. Schlager, 1993). I identified in the curves of tectonic subsidence several pulses of fast subsidence, and I interpreted each of these in terms of tectonic or eustatic processes. Only those pulses of fast tectonic subsidence correlatable to normal fault activity were related to lithosphere extension, while the slower subsidence that followed them was linked to thermal re-equilibration (cooling and contraction) of the lithosphere after stretching. Horizontal extensional movements responsible for the observed subsidence were calculated through forward modelling, assuming that lithosphere stretching produced the Mesozoic subsidence. Using a lithosphere extension model, which considers a two-layers lithosphere (Royden and Keen, 1980), I calculated crustal and subcrustal mantle lithosphere factors that best fit the observed tectonic subsidence signal. The geometry of the basin is reviewed in the context of spatial variations in extensional factors and the location of faults limiting the extensional basin, which controlled subsidence. Basins produced by lithosphere stretching usually are associated with magmatism; therefore, I have also summarized magmatic activity during the Mesozoic and discussed possible relationships between lithos phere stretching and magmatism in the study area. Chapter 3 summar ises the Palaeogene stratigraphy and studies the contemporary vertical and horizontal compressional movements that occurred in the Colombian basin during the Palaeogene. According to some authors ( e.g. Cooper et al., 1995; Villamil and Restrepo, 1997) at that time the Colombian basin was a foreland basin. Other authors ( e.g. Van der Hammen, 1961; Roeder and Chamberlain, 1965; Restrepo-Pace et al ., 1999a,b; Gomez et al ., 1999) have suggested the possibility of upthrusted blocks and/or incipient inversion of the Mesozoic extensional basin in the area of the EC. I discussed and tested these alternative hypotheses using flexural models. I calculated the tectonic subsidence through backstripping techniques. Several hypotheses about tectonic vertical movements are tested through 1D-thermal subsidence modelling and 2D-flexural modelling to explain the observed tectonic subsidence. Because Mesozoic lithosphere stretching generated the basin, the first hypothesis tested (like tested in the Pyrenees by Deségaulx et al., 1991) was whether some remaining thermal subsidence induced by Mesozoic stretching affected the Palaeogene evolution. The second hypothesis tested was one of lithosphere flexure produced by a topographic load in the palaeo-Central Cordillera, as proposed in literature ( e.g.C ooper et al., 1995). The third hypothesis tested was flexural subsidence produced by some local uplifted blocks within the study area. This hypothesis was based on evidence of active deformation, slight inversion and uplift in the hanging-walls of Mesozoic extensional faults in the Colombian basin during Palaeogene. Although the observed subsidence could not be explained by any individual hypothesis, a combination of all gave a satisfactory result. The observed local flexural component of subsidence is interpreted as produced by uplift of local faulted blocks. To better understand the relationships between vertical and horizontal movements, horizontal compressional movements responsible for uplift of local faulted blocks were calculated through forward modelling of structural deformation kinematics, sedimentation and erosion, assuming an incipient inversion of the Mesozoic extensional basin during the Palaeogene (Sassi et al., 1998). This model allowed an estimate to be made of the amount of basin inversion and shortening that occurred during Palaeogene. Chapter 4 su mmarises the Neogene stratigraphy and investigates the vertical movements in the study area during the Neogene, when the Mesozoic Colombian basin was then inverted, deformed and uplifted to form the EC. This study was accomplished by studying the basin subsidence of the Llanos and Magdalena Valley basins using the known sedimentary record as an indicator of subsidence. These
4
Chapter 1
basins became independent sedimentary depocenters in Neogene times. The hypothesis that tectonic subsidence of these basins resulted from lithosphere flexure produced by the increasing topographic load of the uplifting EC during Neogene time is tested through modelling. The potential contribution of Neogene thermal subsidence following Mesozoic stretching was also tested. Chapter 5 giv es an overview of the structural geology of the EC and studies the horizontal compressional movements that were responsible for basin inversion and uplift of the EC during Neogene Andean deformation through map view structural restoration of the EC and neighbouring areas. I manually restored in map view a mosaic of fault bounded blocks using all the kinematic constraints available in the literature, such as strike-slip displacement estimates and amounts of shortening from available balanced cross-sections. This technique is useful to constrain estimates of shortening and strike-slip displacements as well to detect strike-slip motions and rotations about vertical axes not revealed by balancing cross-sections. Chapter 6 pre sents rheological models of the EC lithosphere and based on previous results of the thesis provides a synthesis of the lithosphere scale Meso-Cenozoic tectonic evolution of the EC in terms of stresses affecting the lithosphere and lithosphere rheololgy. Additional ly Chapter 6 pro vides a synthesis of the geological and limited geophysical data, previous results of this thesis and makes comparisons with similar mountain belts and analogue and numerical models from the literature, leading to a hypothesis on the deep structure of the EC.
5
.
s i s e h t o p y h e h t o t g n i d r o c c a e m i t c i o z o s e M e c n i s a r e l l i d r o C n r e t s ) a E 5 9 e 9 h t 1 . f l , o a n t o e i t r u e l o p v o e o c C i n m o r o t c F e ( t . e 5 h t 9 9 g 1 n l . , i w a o t h e s r n e o p o o t r o a C C y : b 2 . d 1 e e s o r u p o g r i F p .
CHAPTER 2 MESOZOIC RIFTING HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES 1. INTRODUCTION Plate boundary forces, deviatoric tensional stresses developing over zones of upwelling asthenospheric convection and frictional stresses on the base of the lithosphere caused by the convective asthenosphere are all driving mechanisms for lithosphere extension (Ziegler, 1994). Therefore, lithosphere stretching events responsible for development of extensional basins should in principle correlate in time and space with periods of increased tensional stress resulting from platetectonic interaction ( e.g. Janssen et al., 1995). The main proposed geotectonic scenarios for rifting are: (1) Atlantic-type rifts developed during the break-up of major continental plates ( e.g. Ziegler, 1993); and (2) Back-arc rifts developed during decrease of the convergence rate and/or divergence of colliding plates (Cross and Pilger, 1982; Tamaki and Honza, 1991). Many features of extensional basins, such as their associated magmatism, geometry and structural style, are controlled by the rheological structure of the lithosphere, the presence of pre-existing faults favourable to be tensionally reactivated and the amount of extension (Ziegler, 1994). The extensional basins developed during the Mesozoic in the area of the Colombian EC ( Fig. 2.1) are good examples to study these processes. However, many features of these extensional basins and their underlying mechanisms are practically unknown. SERRANIA DE PERIJA
1400
S A C U L N A S E D I A N A R R E S
A E N D A L V ) ( M S A G Y T H I L L M E O L E L L Y F O D A L E ID V V A L N A L E M D A O
1300
1200 A R E L N I IL S A B D B R U A S N O L E A A D D C N G O A L H M A R T T N O N D S I E A R - B A C G I R S U B L E Y
1100
1000
900
800
N I A S V I A B B E N U S
800
MARACAIBO BASIN
S D E N A A D I R M E
S A N T A N D E R M A S S IF
S A
F L O M R A E S S S T I F A
F N L T O A R N E D S E T R A H IG H
N I O E G R L S A X I I C U L L A N D O T H N S I N A O S I J O A M F O U N G A M A R B A T S C B O S O A S U A N S U L L Y B - E B A D C U S I A T A N C O N B A G O I O N S A B O E G R
ARAU CA REGION
G Z I N M A B L A A S T A B - B S U
CASANARE REGION
L A ) L ( I N A S B L E S A T N R I E O S N O A L L
L A F I V S S A A M N E E L A A M T D E G A Q U M R E P P I F U S S
META REGION
GUYANA SHIELD
M A O N Z R G A
0
100
1000
200 Km R
E
S
900
BARINAS BASIN
1100
1200
1300
1400
1500
Figure 2.1: Location map of study area with geographic regions referred in the text. Plane coordinates in kilometres are referred to a local origin at Bogota (X=1.000, Y=1.000), as used by the Colombian Geographic Institute IGAC .
Mesozoic Rifting of the Eastern Cordillera, Colombian Andes
In this chapter I investigate the tectonic basin formation mechanisms of the Mesozoic extensional basin developed in the area of the EC. I have studied the tectonic subsidence signals that give important information about basin formation mechanisms. For this purpose, I analysed temporal and spatial basin subsidence patterns, quantitatively analysed tectonic subsidence and forward modelled it, to explain them in the framework of geodynamic processes which formed the Mesozoic EC basin. In doing so I will address questions such as the relationship between basin development, extensional episodes, platetectonic events, magmatic events and basin geometry. Fabre (1983a,b, 1985b, 1987) and Hébrard (1985) studied the subsidence of the eastern flank of the EC during the Cretaceous. They identified the basin as being produced by lithosphere extension, calculated tectonic subsidence curves and, using the uniform instantaneous stretching model developed by McKenzie (1978), calculated lithosphere stretching factors close to 2. They distinguished an Early Cretaceous phase of subsidence produced by rifting and Late Cretaceous subsidence produced by thermal decay after rifting. This chapter studies the tectonic subsidence assuming several events of lithosphere stretching of finite duration and examines the possibility to differentiate between crustal and subcrustal stretching, which occurred in the Colombian basin throughout the Mesozoic. An extensive data set of more than 100 stratigraphic columns and wells from the EC, MV and LLA areas (Fig. 2.2 and Table 2.1) extracted from literature plus well data from Ecopetrol were used.
2. TECTONIC SETTING 2.1. PLATE-TECTONIC INTERPRETATIONS 2.1.1. Tr iassic and Ju r assic
According to plate-tectonic reconstructions (Pindell and Dewey, 1982; Burke et al., 1984; Duncan and Hardgraves, 1984; Ross and Scotese, 1988; Jaillard et al., 1990; Pindell and Barret, 1990; Pindell, 1993, Pindell and Erikson, 1993; Meschede and Frisch, 1998), North and South America were part of the supercontinent Pangea during Triassic time. For the evolution of this period in Colombia the plate-tectonic interpretations rely on two hypotheses ( Fig. 2.3): (1) Intra-continental rifting related to the break-up of Pangea (Pindell and Dewey, 1982; Jaillard et al., 1990) occurred during Triassic and Early Jurassic time (Fig. 2.3a). This hypothesis probably is more applicable to the northern part of Colombia and Venezuela and to their separation from North America. (2) Back-arc rifting behind a subduction related magmatic arc (Bourgois et al., 1982; Maze, 1984; McCourt et al., 1984; Pindell and Erikson 1993; Pindell and Tabut, 1995; Toussaint, 1989, 1995a,b; Meschede and Frisch, 1998). According to this hypothesis, the study area was located at the margin of the continent, when active subduction of oceanic Pacific plates was occurring ( Fig. 2.3b). Such an interpretation explains the Triassic and Early Jurassic rift basins in the study area as backarc basins. The existence of a magmatic arc made up of calc-alkaline plutonic bodies with batholiths of Jurassic age in the area of the Central Cordillera (Aspden et al., 1987) supports the interpretation of back-arc rifting. If the nature of the Triassic-Jurassic plutonic bodies of the Santander Massif (Ward et al., 1973; Restrepo-Pace, 1995) is that of a magmatic arc (Restrepo-Pace, personal communication), then the Santander Massif can be considered as a minor arc segment compared with the longer Central Cordillera arc. Intrusion of magmatic bodies in this small Santander arc is favoured by the existence of the palaeo-Bucaramanga fault-zone (Kammer, 2001). According to Smith and Landis (1995) some arcs migrate either toward or away from the trench in response to changing subduction angles and other processes. Because of arc migration, a single site may change among fore-arc, intra-arc and back-arc settings (Smith and Landis, 1995). Most of the basin may be considered as backarc (relative to the Central Cordillera arc). In addition the portion of the basin located in the Santander Massif minor arc segment may be described as intra-arc (descriptive term sensu Smith and Landis, 1995). Similar difficulties in classifying convergent-margin basins have been discussed by Jordan and Alonso (1987), Flint et al., (1993) and Smith and Landis (1995).
8
Chapter 2 13 SERRANIA DE PERIJA S A C U L N A S E D IA N A R R E S
1400
1300
W
F i g . 2 . 2 4 a
1200
F i g . 2 . 2 4 b
MARACAIBO BASIN
E S D N A D A I R E M
CUCUTA
0
BUCARAMANGA
BARINAS BASIN
100
E
200 Km
ARAUCA
MEDELLIN
A R E L L I D R O C
1100
2 TUNJA
YOPAL
F i g . 2 .2 1
LLANOS ORIENTALES BASIN (LLA)
MANIZALES
BOGOTA
1000 IBAGUE
F i g . 2 . 23 a F i g . 2 . 2 3 b
VILLAVICENCIO
900
5
L A R T N E C NEIVA
800
GUYANA SHIELD S
800
900
1000
SAN JOSE DEL GUAVIARE
1100
R
E
12 1200
1300
1400
1500
Figure 2.2: Location of stratigraphic columns and wells (see Table 2.1) and stratigraphic regional sections. Numbers along sections refer to labelling of stratigraphic transects ( Figs. 2.7, 2.8 and 2.9). Lines E and W denote location of stratigraphic longitudinal sections (Figs. 2.10 and 2.11)
According to Pindell and Erikson (1993), in addition to continued subduction of Pacific plates in the western margin of South America, separation of North and South America started during the Middle Jurassic. As a result a new proto-Caribbean oceanic basin began to open between northwestern South America and the Chortis and Yucatan blocks (Fig. 2.3b). Calc-alkaline intermediate volcanism registered in the Middle Jurassic Saldaña Fm in the south of the basin records arc-related magmatism (Bayona et al.,1 994). For the Cretaceous there are basically three alternative hypotheses (Fig. 2.4): (1) Backarc rifting (Toussaint and Restrepo, 1989; Bourgois et al., 1982; McCourt et al., 1984; Fabre 1987; Cooper et al., 1995; Meschede and Frisch, 1998). Key evidence for this hypothesis is the existence of a subduction-related magmatic arc ( Fig. 2.4a). However, while, there are Cretaceous plutonic rocks in the Central Cordillera, Cretaceous plutonism is only sporadically developed in the northern part of the Central Cordillera, it is practically absent in southern Colombia and Ecuador. It is very extensive in Peru (Cobbing and Pitcher 1972 in Aspden et al., 1987). Aspden et al . (1987) suggested that oblique convergence and an offset in the trench along a major dextral transform fault could explain the notable absence of Cretaceous plutonism in southern Colombia and Ecuador.
9
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes No.
STRAT. COLUMN
1 2 3 4 5 6
Abejorral Alpujarra Apulo Arcabuco Ataco Caqueza
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Cocuy Chima Chita Cimitarra R Luisa Coello El Sudán Floresta Fusagasugá Girardot Melgar Guaca Guataquí Guateque Itaibe La Calera
23 24 25
Labateca Medios Santos
26 27 28 29
Matanza Mojicones Nazareth Neiva
30 31
Onzaga Ortega
32
Pajarito
33 34 35 36 37 38 39 40 41 42 43 44 45 46
Aguazul Paz de Río Prado Q Calambe Q El Cobre Q Olini Quipile R Cusay R Saldaña San Lucas San Pablo Servita San Felix San Luis de Gaceno
47
Simacota
REFERENCE Burgl and Radelli (1962); Botero-Arango (1963) Mojica et al. (1985); Salazar (1992); Florez and Carrillo (1994) Cáceres and Etayo-Serna (1969); Etayo-Serna (1979); Sarmiento (1989); Martínez (1990). Renzoni (1967); Renzoni et al. (1967) ; Galvis and Rubiano (1985) Cediel et al.( 1981); Geyer (1982); Mojica (1982); Mojica and Llinas (1984) Hubach (1957); Julivert (1962, 1968) Renzoni (1965a,b, 1968); De La Espriella and Cortes (1982); Pimpirev et al.( 1992) Etayo (1985); Fabre(1985a,1986,1987); Mayorga and Vargas (1995) Julivert (1958a,b); Pulido (1979); Pulido et al.( 1979a,b, 1980); Rolon and Carrero (1995) Fabre(1985a,1986,1987) Alfonso and Ballesteros (1987); Alfonso (1989) Tellez and Navas (1862); Mojica et al.( 1985) Tellez and Navas (1862); Mojica et al.( 1985) Geyer (1976, 1982) Mojica and Villarroel (1984); Alzate and Bueno (1994); Alzate and Rolón (1996) Laverde and Ramírez (1987); Laverde (1989) Burgl and Dumit (1954); Raasvelt (1956) Burgl and Dumit (1954), Raasvelt (1956); Diaz (1994a) Maughan et al.( 1979); Fabre (1985a, 1986, 1987) Porta (1965); Julivert (1968); Schammel and Buttler (1981, in ESRI - ECOPETROL, 1994) Burgl (1958); Ulloa et al (1975); Ulloa and Rodríguez (1976a); Mayorga and Vargas (1995) Mendivelso (1982) Van der Hammen (1958); Ujueta (1961); Renzoni (1965a); Julivert (1968); Cuervo and Ramírez (1985) Carillo (1982); Boinet et al.( 1982,1985) Renzoni (1967); Renzoni et al.( 1967); Diaz and Sotelo (1995) Julivert (1958); Zamarreño de Julivert (1963); Julivert et al. (1964); Cediel (1969); Laverde (1985); Alfonso (1985) Julivert (1959) Daconte and Salinas (1980); Fabre (1981,1987) Fabre (1987) Burgl (1959); Beltran and Gallo (1968); Howe (1969); Wellman (1970); Anderson (1970,1972); Waddell (1982) Daconte and Salinas (1980) Burgl (1961); Cediel et al. (1981); Mojica et al. (1985); Carrillo and Flórez (1994); Díaz (1994); Amezquita and Montes (1994) Miller and Etayo-Serna (1979); Hebrard (1985); Naar and Coral (1993); Cardona and Gutierrez (1995); Mayorga and Vargas (1995) Miller and Etayo-Serna (1979); Hébrard (1985) Alvarado and Sarmiento (1944); Walthill and Berry (1979); Guatame and Lara (1995) Vásquez and Ríos (1979); Mojica and Macía (1981); Salazar (1992); Díaz (1994a) Flórez and Carillo (1994) Barrero (in Julivert, 1968) Villamil and Arango (1998); Flórez and Carrillo (1994); Villamil (1994) Sarmiento et al.( 1985); Sarmiento (1989); Martínez (1990); Gómez and Pedraza (1994) Ecopetrol(1970); Fabre (1987) Cediel et al.( 1981); Geyer (1982); Mojica (1982); Mojica and Llinás (1984) Bogotá and Aluja (1981); Geyer (1982) Hall et al.( 1972); Etayo-Serna et al.( 1976) Julivert (1960); Vargas et al. (1981ª ,b, 1984a); Fabre(1987) Etayo-Serna (1985); Rodriguez and Rojas (1985) Ulloa et al.( 1975,1976b,c); Ulloa and Rodriguez (1976a); Robertson Reseach Inc. (1983, 1986) Pulido (1979a,b, 1980); Rolon and Carrero (1995)
Table 2.1: Stratigraphic columns used in this thesis
10
Chapter 2 No. 48 49 50 51 52 53 54 55 56 57 58 59
STRAT. COLUMN Simijaca Sogamoso Sutamarchan Tabio Tablazo Lebrija Tibasosa Tunja Chivatá Vado Real Vélez Villa de Leiva
60 61
Villeta Yacopí
62 63 64 65 66 67 68 69 70
Apiay-4P Bolivar-Corrales Cormichoque-1 Chichimene-4 Chitasugá-1 Cusiana-1X-2 Leticia-1 Medina-1 Sabalo-1K
71 72 73 74 75 76 77 78 79, 80 81 82 83 84 85, 86 87 88, 89 90 91 92 to 102 103 104 to 107 108 109 to 124 125 126 127, 128 129 to 132 133 134 to 139 140 141
Llanito-1 Casabe-199 Infantas-1613 Cascajales-1 Suba-2 Suesca-1 Suesca Norte-1 Tunja-1 Wells Llanos Arimena-1 Balastrera-1 Cachama-1 Camungo-1 Wells Llanos Caño Cumare-1 Wells Llanos Caño Limón-1 Caño Rondón-1 Wells Llanos Joropo-1 Wells Llanos La Tortuga-1 Wells Llanos San Juan-1 ST-0-04 Wells Llanos Wells Llanos SV-9 Wells Llanos Paime Chenche-1
REFERENCE Ulloa and Rodríguez (1979, 1987, 1991) Mayorga and Vargas (1995) Ulloa and Rodríguez (1979, 1987, 1991) Burgl (1958); Julivert (1962) Morales et al.( 1956); Cediel (1968); Ward et al (1973); Renzoni (1985); Clavijo (1985) Morales et al.( 1956); Cediel (1968); Ward et al (1973); Renzoni (1985); Clavijo (1985) Renzoni (1967); Renzoni et al. (1967); Alzate and Bueno (1994) Renzoni (1967); Renzoni et al.( 1967) Renzoni (1967); Renzoni et al.( 1967) Pulido (1979a,b, 1980) Ulloa and Rodríguez (1979) Etayo-Serna (1968), Galvis and Rubiano (1985); Cardozo and Ramírez (1985); Villamil (1994); Villamil and Arango (1998) Cardozo and Sarmiento (1987); Sarmiento(1989); Martínez (1990); Moreno (1990) Moreno and Rubiano (1987); Rubiano (1989); Moreno (1990a); Rodríguez and Ulloa (1994a,b) Ecopetrol and Beicip (1995) Well log; Céspedes and Peña (1995) Well log Ecopetrol and Beicip (1995) Well log Fajardo et al.( 1993) Fajardo et al.( 1993) Fajardo et al.( 1993) Ecopetrol et al.( 1994); Ecopetrol-ICP (1996); Ecopetrol, Vicepresidencia de Exploración, (1997); Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Well log Well log Well log Well log Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Ecopetrol and Beicip (1995) Fajardo et al.( 1993) Rubiano (1989) Ecopetrol, Vicepresidencia de Exploración, (1997)
Table 2.1: Continued
11
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes No. 142 143 144 145
STRAT. COLUMN Valle de San Juan Cambao Simiti Palmichal
146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 to 162
Tausa Honda Sabana de Bogotá Chicoral-1 Sapo-1 Suárez-1 Venganza-1 Revancha-1 Manuelón Q Michía Desengaño Palogrande Campoalegre Q Guadinosita Wells Llanos
REFERENCE Barrero (in Julivert, 1968) Schamel and Butler (1981 in ESRI – ECOPETROL, 1994) Dickey (1941); Morales et al.( 1956); Etayo et al.( 1976) Ulloa et al. (1975,1976b,c); Ulloa and Rodriguez (1976a); Robertson Reseach Inc. (1983, 1986); Bartels (1986); Sarmiento (1992, 1993) Follmi et al. (1992), Sarmiento (1992, 1993) Porta (1965, 1966); Julivert (1968); Van Houten and Travis (1968) Helmens (1990) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Ecopetrol, Vicepresidencia de Exploración, (1997) Howe (1969, 1974); Wellman (1970) Howe (1969, 1974); Wellman (1970) Howe (1969, 1974); Wellman (1970) Howe (1969, 1974); Wellman (1970) Howe (1969, 1974); Wellman (1970) Howe (1969, 1974); Wellman (1970) Fajardo et al.( 1993)
Table 2.1: Continued 2.1.2.
Cretaceous
(2) Passive margin (Pindell
and Erikson 1993; Pindell and Tabut, 1995). The scarcity of magmatic rocks in the basin seems to support this hypothesis ( Fig. 2.4b). However, the presence of some plutonic rocks of Cretaceous age in the Central Cordillera is difficult to explain with this hypothesis. (3) Intracontinental rifting related to the break-up of Pangaea. Some authors ( e.g. Geotec, 1992) following this hypothesis suggested a NW-SE graben developed in the northern Central Cordillera during Early Cretaceous. The presence of some plutonic rocks of Early Cretaceous age in the Central Cordillera ( i.e. San Diego, Altavista, Cambumbia and Mariquita stocks; Restrepo et al ., 1991; Toussaint and Restrepo, 1991) interpreted as part of a subduction-related magmatic arc (Aspden et al ., 1987) is difficult to explain with this hypothesis. Most plate-tectonic assume that during Early Cretaceous western Pangaea was already broken ( Fig. 2.4). During the latest Cretaceous (post-Santonian) all plate-tectonic interpretations propose a convergent margin west of Colombia. The Caribbean plate was moving eastward relative to South America, while the Farallon Plate (Fig. 2.4) was subducting west of Southern Colombia (Pindell and Erikson, 1993 and Pindell and Tabut, 1995).
3. STRATIGRAPHY 3.1. TRIASSIC AND JURASSIC SYN-RIFT SEDIMENTATION The Triassic and Jurassic sedimentary record is present in several isolated outcrops ( Fig. 2.5). Continental deposits with red beds and volcanic effusive and pyroclastic deposits are dominant, although there are locally some marine facies. Triassic and Jurassic rocks were deposited in rift extensional basins mainly located in the Upper Magdalena Valley, Serrania de San Lucas and the western flank of the EC (Mojica et al.,1 996). Figure 2.6 shows a stratigraphic synthesis modified from Mojica et al. (1996, their Fig. 2.6). Triassic and Jurassic sedimentary rocks formed a sequence bounded by unconformities. The lower contact is marked by an unconformity, which is dominantly angular. The upper contact is dominantly unconformable but locally it is conformable. Jurassic deposits consisting of clastic facies deposited in dominantly continental environments are widely distributed. In those sections where there are some marine facies, they are underlain and overlain by continental clastic facies. The fine-grained muddy marine facies record local marine incursions. Volcaniclastic, pyroclastic and volcanic lavas are mainly restricted to the upper part of the Upper Triassic to the lower part of the Middle Jurassic (Mojica et al.,1996; their Fig. 2.6).
12
Chapter 2
A
Y
c u
Y
S P C E C C
C E C C
S
i s t r h o C
N
P
G S
1
i s r t h o C
B P
5
J
S J
C E
N n a
t s r i h o C
t
u Y
A
C E C C
S
C C
i s o r t C h
a c
u Y
P
13
R
1 S
; i a l a 0 r b t 8 9 n m e 1 l o C , o y e C o e w a e f o l a D t P s e C d n a w C s , l e l n t e i a s d a n l p b i c l P i a n n m a o o e i r c f ( o s n e c t i g x f i n e i c t a a f r i r P l e f l l i a o t d r n n o e o i n t C i t c n u n r o d e t c b s a u a r s E t n o i t o f e o d a e l t s a i a l s P e e : r h t C g o n E p . t ) y i f H i r 5 . c 9 9 1 A r a , t k c : u a b e b a m f T i t o c i s d i n s a s s l e a l h r e u t d o J n p c y i i P s H ; s a . 3 i r 9 . 9 ) T B 1 1 . . r n 2 o ) 0 o . f 9 s i g s 9 k i n r F o 1 ( , . E i t l a c a d e u n t r e a r a t s l y l d n r e d o u a d c l l i n t s e r i a P f J c ; o i ; 0 n n 8 o o 8 9 i t 9 t 9 c 1 e 1 . , a c t , l l o e e a t s : a t s t l t e e P e s o d n r I c : . S l a . 3 l . d i a r 2 a l e n J l a e i r s m d u o g s o f o r r i F R ( C
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
C E C C
P
C E C C
e i m a m s - A a l l i t A n
S
e a i m m A s t i l la A n
N
B
S
C
C
5
E
S C S E
C C
G
M C E C C
N
Y
C
C
A F
14
n l a i r g t r a n e m C e v o e i a s l s a a P P . C B C , . n ) i 8 s 9 a 9 1 b l , . a l n a o t i e s n e e t d x e e h c a s r e l e M l i d r m o o r C f ( n s r i e s t e s a h t E o p o y e h l a a g n P i t : f C i r E c r a . k ) c a 5 B 9 9 . 1 , A t u b : e a m T i t d n s a u o l l e e c a d t n ) . e r i 1 P . C ; 2 r 3 . o 9 i g f 9 s 1 F ( n o n a i e t o r s c a k u i r r y t s E d n u t o d s c n f e a r o c l l n i e i o n d t o n a t i c c e P l o t : e m t t o a r e s l f n P ( I . . s 4 a . i r s 2 e e l h l e t i r o d u r g p o i y C F h
Chapter 2
Cosinas Gp Cojoro Gp
Corual Fm
Guatapuri Fm
ay t) u nd r u s o B l C a m n t s te in e e n t ( W C o
n b e r a G o a r z L a n a S
n b e La Quinta Fm r a G k e a l L r a t e n C
h g o u r T s e u i q h c a M
Los Indios Fm
Morrocoyal Fm ) 5 9 9 1 t, in a s s u o ( T e n a rr e T i m a h a T
s a u c L n a S d e ia an rr e S
Serrania de Perija
Catatumbo La Quinta Fm region BARINAS BASIN
Giron Gp
W N IO T C E S
h u g o r T e n t SE CT I O N a 1 b i r U
Triassic-Jurassic Arauquita-1 Matanegra-5 and Guafita-5X wells
Rusia, Montebel and Palermo Fms
r e t u u S l r a e m o R
LLANOS ORIENTALES BASIN (LLA) Bata Fm Brechas de Buenavista Fm
Payande Gp
E N O T I C E S
GUYANA SHIELD
0 -
100
200 Km
Study area (Fig. 2.1) Outcrops of Triassic and Jurassic strata (Maze, 1984; Toussaint, 1995) Boundaries of Triassic and Jurassic troughs (Julivert, 1968; Maze, 1984)
Motema Fm
TRIASSIC-JURASSIC BASIN COMPARTMENTS (Refered in Figure 2.20 ) Payande, San Lucas and Sierra Nevada Terranes acording to Etayo-Serna et al (1986), western part of Chibcha Terrane (Toussaint, 1995). Eastern part of Chibcha Terrane (Toussaint, 1995) and Guyana Shield.
Figure 2.5:Location of Triassic-Jurassic strata outcrops and stratigraphic sections. Labeling of stratigraphic sections according to Fig. 2.6. The stratigraphic section E represents the Triassic-Jurassic sedimentary record of the eastern part of the Chibcha Terrane according to Toussaint (1995). The stratigraphic section W represents the Triassic-Jurassic sedimentary record of the western part of the Chibcha Terrane equivalent to the Payandé, San Lucas and Sierra Nevada Terranes of Etayo-Serna et al. (1986). (modified from Toussaint, 1995). Inset: location of Figs. 2.1 and 2.2.
15
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
. W S S
z n a l a P
a n a a p t u i z j i u a U C C
a r i j a u G
a t n i u Q a L
o l r e t e h C
F a c O
W S S
e d n a r G o h c n a R
n a p t i U
z n a l a P
l a u r o C
i r u p a t a u G
a d a r r a v e e i S N
a t n i u Q a L
s o i d n I s o L l
n a a y e o c r o o r N r o M a n a j o M a L
n a d u S l E
n a d u S l E
) e n a r r e T a h c b i h C f o t r a p n r e t s e w (
s n a a c u S L
) e n a r r e T a h c b i h C f o t r a p n r e t s a e (
s a c u L n a S
n o r u b i T
a g a z n O
o c u b a c r A
o c u b a c r A
a l s u R a L
l e b e t n o M
o m r e l a P
o l v a u G . z a C e t s i m v a a ñ n a e u S B
a t a B
o l l e o C
o l l e o C e r b o C l E . Q
a o l d a c a i r P h C o o b b M M e d n a y a P
o c a t A a ñ a d l a S o i R
a v i e N
a s i u L
a v i e N
a m e t o M
a ñ a d l a S
e d n a y a P Y L R A E
E T A L
S U O E C A T E R C
) a M ( e g A
s o t n a n n s o S i r a a c r s G d o o o J B L
o z a l b a T
E N O I T C E S
W N O I T C E S
E N N
e r b m u C
0 . 2 4 1
E L D D I M
C I S S A R UJ
E N N
Y L R A E
E T A L
E L D D I M
Y L R A E
) a M ( e g A
C I S S A I RT
7 . 5 0 2
E Y Y L E L L D T R R A D A A I L E E M C I S S A R UJ S U O E C A T E R C
2 . 8 4 2
16
0 . 2 4 1
E T A L
E Y L L D R D A I M E
C I S S A R I T
7 . 5 0 2
2 . 8 4 2
l a h t o t i n w h o r z e t i o c r n n o ( e h r a e e d f n r i a . a f l ) e d n a 3 . I m y i . o 7 t ) c l 6 o n a o r c 9 i r i 9 t o g c o 1 , . M e l s l o a e n e i d s g t s e a n t a o n c g i a i e t j n s a o t é e i i r M p C m e l r m e i a s o h t i t r a f x n d a d i l e . e a e f c i s i m ( i t d F r e o n a r m á e V . ( d a 5 u é . n d o S 2 i n t . a l a g r E y i o a F t s P u e n s e i r o r h s e t c f t i f i t c l s o e i a s . s s p n m s a o r i n f F t l f y a s o a i r p l u o t n o L o u i o p e t h a e h t h t c i o w t h L ) t t . a i s m h t w n e k o ( d i v i t e e t c t a c l s e s n e e a r g c t r s i g o i h u d s c p l d a a s r n a g t a n i h t o c a i ) z r i s t 2 r s o 8 s a i c h 9 r i 1 s ( T y s a r i s a d r e ) u t y J e s n a e c c G s i u e s r n s o L a p i i n r s t a a S T t t n e e : s r e p 6 d e . r r 2 p t e í a e e n n r i r a u s i s r r g i i e F x a h t S
Chapter 2 SECTION 1 WNW
ES E
Western Chibcha Terrane
Ag e (M a)
142.0
La Cumbre
Magdalena palaeo fault system
C I L A T E S S A RM I D D L E U J EARLY
205.7
248.2
Bituima La Salina palaeo fault system o s z o a t l b n a a T S
n a d u S l E S U O E C A E A R L Y T E R C
Eastern Chibcha Terrane Guaicáramo palaeo-fault system
Arauquita-1 Matanegra-5 and Guafita-5X wells
Macanal
L os Santos Giron Jordan Bocas Tiburon
La Mojana Morrocoyal
C I L A T E S S AM I D D L E R I T E A R L Y
Guyana Shield
Corual
0
100
200 Km
DEPOSITIONAL ENVIRONMENT Volcanic and volcaniclastic rocks Alluvial fan an d fluvial sandstones and mudstones Coastal plain and lacustrine mudstones and sandstones Littoral to inner shelf sandstones Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltstones Outer shelf shales or carbonates Turbidites sandstones and mudstones
Figure 2.6: Continued
Facies and thickness similarities related to their geographic position suggest that TriassicJurassic sedimentation occurred in two separate basin compartments, each one with its own subsidence history and sedimentary fill ( Figs. 2.5 and 2.6): (1) Upper Magdalena and Cienaga de Morrocoyal (region A in Fig. 2.5). It corresponds with the western part of the Chibcha Terrane as defined by Toussaint (1995a,b) or the Payande and San Lucas Terranes proposed by Etayo-Serna et al., 1983; Fig. 2.5). In this area a marine incursion with a correlatable maximum marine flooding surface is recognisable near the Triassic-Jurassic boundary (Chicala member of the Saldaña Fm in the UMV, the Morrocoyal and Los Indios Fms, ages according to Mojica et al ., 1996; their Fig. 2.6). Continental sedimentation followed by shallow marine limestone deposition occurred d uring Triassic (Norian?-Rhetian) in the Upper MV. Volcanic-related facies are volumetrically more important in this area than in the Eastern Cordillera area (Region B in Fig. 2.5). (2) Eastern Cordillera (Region B in Fig. 2.5). This corresponds to the eastern side of the Chibcha Terrane as defined by Toussaint (1995a,b; Fig. 2.5). In this area a marine incursion with a correlatable maximum marine flooding surface is located within th e Lower Jurassic (upper part of Bata and Montebel Fms, ages according to Mojica et al ., 1996; their Fig. 2.6).
17
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
3.2. CRETACEOUS SEDIMENTATION The majority of exposed rocks in the EC are Cretaceous in age . Figures 2.7 to 2.11 show two longitudinal and 3 traverse time-stratigraphic cross-sections of the basi n. Location of these sections is shown in Figure 2.2. They have been constructed from earlier versions by Fabre (1985a, 1986, 1987) and Coo per et al (1995), modified according to sequence stratigraphy interpretations from literature (Pimpirev et al ., 1992; Fajardo et al ., 1993; Etayo-Serna, 1994; Ecopetrol et al., 1994; Villamil, 1994; Rolón and Carrero, 1995; Villamil and Arango, 1998). Cretaceous rocks, including locally the uppermost Jurassic and Paleocene deposits, form a mega-sequence bounded by regional unconformities that are at least locally angular. On a broad scale, Cretaceous rocks represent a major transgressive-regressive cycle with the maximum flooding surface close to the Cenomanian-Turonian boundary, corres ponding to the maximum Cretaceous, and even Mesozoic, eustatic level (Fabre, 1985a; Villamil 199 3; Figs. 2.7 to 2.11). Superimposed on this largescale trend, several smaller transgressive-regressiv e cycles are present, suggesting an oscillating relative tectono-eustatic level. Subsidence was rapid ( Fabre, 1983a,b, 1985b, 1987; Hébrard, 1985), but shallow water sedimentation suggests that deposition kept pace with it. The basin was a wide rift system oriented approximately NNE-SSW, divided into two sub-basins (western Tablazo sub-basin and eastern Cocuy sub-basin) by the Santander-Floresta palaeo-Massif. Toward the north, these sub-basins continued to the Machiques Trough in the Mérida Andes of Venezuela and the Uribante Trough in the Serranía de Perijá (Julivert, 1968; Fabre, 1985a, 1987). Toward the south these sub-basins joined as the Cundinamarca sub-basin (Bur gl, 1961), where the thickness of the Cretaceous sections reaches a maximum (Figs 2.1, 2.12 and 2.13). Fabre (1987) and Sarmiento (1989) suggested that the Cundinamarca Trough was limited to the south by a NW-SE transfer palaeo-fault. South of it, there was an extensional relay system (Restrepo-Pace, personal communic ation). N-S lateral changes of thickness suggest the existence of other NW-SE trending transfer fault s (Fig. 2.12 a nd 2.13). Based on the presence of Lo wer Cr etaceous sedimentary rocks in the norther n part of the Central Cordillera (Figs. 2.12, 2.14 and 2.16) Geotec (1992) suggested the existence of a NW-SE grab en in this area, it was connected with the Cu ndinamarca sub-basin. Compiled available thickness of these sediments of the Central Cordillera for similar chronostratigraphic intervals, however, is significantly smaller than those of the EC( Figs. 2.13, 2.15 and 2.17). If such a graben existed, in terms of subsidence it was a minor feature compared with the gr abens in the area of the EC. 3.2.1. Ear ly Cretaceous syn-r if t sedimentation
The Early Cretaceous sedimentary history is illustrated in Figures 2.12 to 2.17. S edimentation started in the Tablazo sub-basin in Jurassic time and continued during Early Cretaceous locally without a tectonic-related angular unconformity ( e.g. at the Rio Lebrija section, Cediel, 1968). In other areas Cretaceous sedimentary rocks rest with an angular unconformity on earlier Mesozoic, Palaeozoic or even Pre-Cambrian rocks. In the Tablazo sub-basin the first facies deposited were mainly sandstones (Los Santos, Tambor, and Arcabuco Fms) deposited in fluvial environments (Renzoni, 1985a,b, c; Clavijo, 1985; Vargas et al., 1985; Laverde and Clavijo, 1985; Galvis and Rubiano, 1985; Etayo-Serna and Rodríguez, 1985). Bürgl (1960, 1964, 1967) suggested that an initial marine incursion in the Cundinamarca sub-basin flooded a continental area with a desert climate, which provided conditions for evaporite formation during the early stages of marine transgression. McLaughlin (1972) cited palaeontological evidence of Berriasian-Valanginian age for some evaporite occurrences. During the Berriasian the sea flooded the basin from the northern part of the Central Cordillera toward the Cundinamarca sub-basin (Etayo-Serna et al., 1976). Then the sea advanced from the Cundinamarca sub-basin toward the north into the two sub-basins while the Santander-Floresta palaeo-Massif remained emergent (Etayo-Serna et al., 1976; Fabre, 1985a, 1987; Sarmiento, 1989; Cooper et al., 1995; Figs. 2.12 to 2.17).
18
Chapter 2
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 1
8 - N C T S
E S E
S E L A T N E I R O S O N A L L
1 a l l e t S
o a r b e r o a n d y n o a r a o b i r u e a M G L C
e p u l a d a u G
1 - o p o r o J
2 N O I T C E S
a t e h c a G
1 - n o d n o R . C
t l u 1 - e l E . R a F o m a r á s u C R c y a i a u G
s o v r e u C o c r a B
y u c o C
s e n o c i j o M
A R E L L I D R O C N R E T S A E
Y E L L A V A N E L A W D N G A W M E L D D I M
s e n o t s s d e n n a o t s s d d u n m a T d s e N n n o E a t s M s e d N n u o m O t s R e I d n V n i r N a s t s E l u a c i L v l a A f u l d N n O I d n a T n I a i a S n l a p O f l l P a a t E i v s D l u a o l A C
a t i v r e S R
n o i c a r t n e c n o C
o h c a c i P
a c a u G
s o t n a S s o L
t l u a F o l b a T a z a n i l a S a L 3 1 6 1 s a t n a f n I
z a P a L s a d l a r e a s m o s a l o r E s a d g e e a u r M M R l o o C
r o b a L
a n r e a i h T c o S c r A a s h c a o u S d a r u A G n a u J o t i M a n n u o L l o a C L
s e n o t s d n a s f l e h s r e n n i o t l a r o t t i L
l o t s u a t m a a f n r u o á J c e n e i n o s a a l U p a l u a A L a G P n t a l c u a á f a t M i v o r e e a l S o a r g P e N t l o s i u e R a á f d c e a o c y r o e e l B a e M a u P q u a b i p i T h C
s e n o t s t l i s d n a s s e e t a n o n t o s d b r u a c m f f l e l e h h s s r e r e n n n i n i e e n i n i r r a a m m w w o l o l l l a a h h S S
N I S A B B U S Y U C O C
a a i n t m i a r u i s m L m i L U a i L S
a o z a c a j n l a a l b P b a a T s o R
t l u z f e a r a o u e a S l a P
t l a u n f a i l a S o e a a L l a P
í t i m i S L M E L E L M E L E Q I O I C L M C
L I L O A P M O E P Y R A I T R E T
) 0 0 0 0 0 a 1 2 3 4 M ( e g A
L E M S T C C
E T A L
A
B T B B A V K C A B H O R V
Y L R A E
S U O E C A T E RC
0 6 0 0 0 0 0 0 0 0 5 7 8 9 0 1 2 3 1 1 1 1
19
f l e h s r e t u O
0 m 0 2 K
0 0 1
n o r i G s o t n a S s o L
s e t a n o b r a c r o s e l a h s
L D D I M E T A L E
C I S S A R UJ 0 0 0 0 0 4 5 6 7 8 1 1 1 1 1
s e n o t s d u m d n a s e n o t s d n a s s e t i d i b r u T
0 O Z A L B N I A S T - A A B N B E L U S A D G A M
e , l a 5 c s 8 e 9 1 h , t e r o t b a g n F i r d e r t f o a c ( c a n o e i t m i r t a o l t a s c e i r g c o i l t s o e p g a s s n t i n l e a s p e t r u p o e r h t s i i w x a ) l m a c k i ( t r e e c n V a . t 2 s . i 2 d e l a r t u n g i o z F i r n o i h n o y a i t d a t c o n e s L e . 2 r p n s t o i n t e c s e s e c r p i e h r p s a i r x g a ) i 5 t l a a 9 r t 9 t s n 1 o , . y z l r i r a a i o t t r e h , e r T ) e 6 p d 9 o n 9 a 1 o ( C s u g ; o g 4 e 9 O c a d 9 1 t e i r n a l C i m n : e l a 7 l t . i s 2 d V e a ; r r 7 u G 8 g 9 i F f o 1
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 1
E S E
5 1 - U G T S
1 - s a n a l P
1 - a i l u r a B . C
S 1 - a n e m i r A E L A T N E I 1 i r a u G R - a n e m O S O o b N a n A y L 1 i r r e r t n E a o L - s o u e
a r r e o n d o r a L b G r i a M C
1 - N a i r o l G
5 N O I T C E S
o c r a B s o v r e u C
1 a í r a M a L
t - a n a i s u C l 1 u a F o m o t i r a j a P a r a c i a u G
a j n u T
A 1 a c m r o C t R - e u q o h a l E 1 - a j n u T i L T L I D R O C a v i e L e d a l l i V N R a r i u q i n o M E T n a h c r a m a t u S S A a c a j i m i S E
n o i c a r o t n h c e c a n i c o P C
z e l e V t l u a F a i p c a Y n o i l a S a L
a t e h c e a p G u l a d a u G
s e n o t s s d e n n a o s t s d d u n m a T d s e N n n o E a t s s M e d N n u o m O t s R d e I n V n i r s t N a s E l a u c L i a v l A l u N f d n O I d n a T a n I i a S n l a p O f l P l a a t E i s D v u a o l A C
s e n o t s d n a s f l e h s r e n n i o t l a r o t t i L
s e t a n o b r a c f l e h s r e n n i e n i r a m w o l l a h S
s e n o t s t l i s d n a s e n o t s d u m f l e h s r e n n i e n i r a m w o l l a h S
t o l m u a a f r a á o r c u i e a a D a l n u a a r e G P s c o a a a b u t h ) a q n M a l c ? L i u p ( o t J l S h u a s C r a a g a A n á f e L r e p o e u n i a e T s U q a o l e r a S a m a h e P o s c n o o F s e S a a l b t P i c l r T u A a á c f a y o o e l B a f a n I P t l l o i u c a l p G u z f e u n b e s a r a f a a a S S a o c e t a r u u l a o d i R l A S g a o a G z o u n n a P B G a a l b S S a a n T j o r a i P a c G t n l a a u l n f a b i l a a s S o o e a R a a i t L l a i m i r a m P a i s m n i u S L U L a L
s e t a n o b r a c r o s e l a h s f l e h s r e t u O
N I S A B B U S Y U C O C
Y R A I R T E T
S U O E C A T E RC
C I S S A R UJ
20
m K 0 0 2
0 0 1
0
O
Z Y A E L L B L N s I A a A T d V - S l A a A A a r B N e o s N o m E B d r E L U a g s L r u E A S o A l D o M - o r e d a t a M a l D 2 G a s e C A G e R M W A M M N E o l W L i D H D I L M E L E L M E L M E L E Q B T M K O C B B A I I O I C L M C S C T C A A B H V L L O A R V Y L R A E L D D I M P M O E P E T A L E T A L E ) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 1 M ( e g A
s e n o t s d u m d n a s e n o t s d n a s s e t i d i b r u T
c n i t i s d a r p o s c n c i a l e a p m i t t u l o a h t c i i g w o ) l o e m g k ( s t e n c e n s a e t r s p i e d r l s a i t x a n o l z a i r c i o t r h e V y . a d 2 . 2 t n e e r s u e r g p i . F t s ) n 5 i n 9 e 9 s n e 1 o r i , . t p a l e c a o r s t e L i . x r 5 a e p l n o a o o t i t n C c o ; e z 4 s i r 9 c o 9 i h 1 h p , l i a ) r 6 m g 9 i a t l 9 l a i 1 r t V s ( ; g y 7 g r a O 8 i 9 t 1 r e d , T n a 5 8 d n 9 n a i e 1 , t s s e r u d b o a a r F e c G r a t f e e r o t f a C l ( e : n a 8 . c o i s t 2 e e a r r h o u t t g i o s F t e r
Chapter 2
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 1 1 1 1 1 1 1 1 1
E S E
T N E M N O R I V N E L A N O I T I S O P E D
9 V S
1 - s a i a c a n a M
S E L A T N E I R O S O N A L L
1 - e n i g a r o V
A 1 - e n e m i h c i h C R t E l L u a L I F D o R m O a r a z e u q a C C á c i
N a u R G a r e l a C a L E T S 2 - a b u S A E
a r r e o o n d b o a a r r y b a a i u C M G a n L o e L
e u q o h e a r c e a m d b s a u g S U e R a n a a L b a S
a l e u h c i t l r u a a t e l l i V a M F e a a d o s n c m i e a e u t a r u s e J i i n a i V a d T B 1 o n i - n a r t l e B s e o t a n R a H M n S a S 2 - a g u s a t i h C o i b a T
W N W
s e n o s t s e t d a u n m o b r d a n c a r s o e s n o e t l s a d h n s a f s l e s h t e s i d r i e t b u r u O T
m K 0 0 2
0 0 1
a t e h c a G
1 - y a i p A
Y E L L A V G A M E L D D I M
s e n o t s d n a s f l e h s r e n n i o t l a r o t t i L
s e t a n o b r a c f l e h s r e n n i e n i r a m w o l l a h S
s e n o t s t l i s d n a s e n o t s d u m f l e h s r e n n i e n i r a m w o l l a h S
1 - y a b i n a G
1 - s o t i r g e N
2 1 N O I T C E S
s e n o t s d u m d n a s e n o t s d n a s l a i v u l f d n a n a f l a i v u l l A
s e n o t s d n a s d n a s e n o t s d u m e n i r t s u c a l d n a n i a l p l a t s a o C
0
e e o r u n h o c b a q U a z a a r a e p u L C i u l t D h . q a a r a s a i C e A n u C n r a v a q e i e c a n T m M e u o B F a s t a o u g d o a B u G s a a r a c r r e e u h t c n a n M a n ó c o i á c r t r i y e T t o F o H S o c o U l i S H t l o u a a f b m o a e a C l a P
L L M E L L E M E E Q
L E M
o t l m u a a r f á c o e i a a l u a P G
t l s u a a f o n o a e a C l a P t l u a f a m i o u t e i l B a a P
B T
V K O A C B B O I C L M C S C T C A A B H I I R V L O A L Y L R A E E L D D I M E T A L E T A L M E P O P S U O E C A T E RC Y R A I R T E T C I S S A R UJ ) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 a 1 1 1 1 1 1 1 1 1 M ( e g A
21
o n t o i g t n a i r d o r t s o e c r c a i c e t a m s i t p s l n a i l c i a g p o t l u o o e h g t i s w t n ) e s m e r k ( p e r e s c n i a x s a t i l d a c l i a t t r n e o v i z 2 . r o 2 h . g y i a F d n t i n e n e o s i r t p a c s o t L n e s 2 e ) . ) 1 r p 5 e n r o s 8 9 i t 1 i c , x e a s l l a c t a t i e h n r p e o a z p r i o g r i o o t h C a r , ; t 4 s ) 6 9 9 y 9 r 9 1 a 1 i l ( t i r e g m T g l a O l d i n V a d n ; a 7 s u 8 n o i e e 9 t c , s 1 a 5 t d 8 e r a r 9 C G 1 , e : f r o 9 . e b a 2 l F a e c r r s t e u g e f i h a F t (
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes 0
0 1
0 2
0 4
0 3
0 5
0 6
0 7
0 8
0 9
0 0 1
0 1 1
0 2 1
0 3 1
0 4 1
0 5 1
6 1
0
0 7 1
0 8 1
a c e t a b a L y a s u C R
s IN S
y u c o C
U C O C
E N N A R E D R O C E T S A E
E N IO T C E S
s
d u m d n a s e n ot
o e al a P
W S S
Y E L L A V A N E L A D G A M R E P P U
s
lul A
a o C s
l v
r
fe s n a
D R O C N R E T S A E N R E
d n a
ai
d n a n af -
s
u c al d n a ni al p l at
f
u af
e n ot
s
s
e h s r
e n ni ot l
s
ra ot ti L
tr
D N U C
A C R A M A NI
NI S A BB U S
tl
u af
-
o e al a P r
fe s n a tr
a z e u q a C
h t e r a z a N
a d n o H
ht e tl r u a z aF a N t o d r a r i G
-
o e al a P ht e r a z a N
a s e M
a B
a M( e g A
0
0 1
L
M
I
IO M
L P
)
L
M
I
m K 00 2
r
fe s n a
00 1
tr
u d a u G L
Q
IN S A BB U S
0
a d n o H
a v i e N
u af
a d n al a u G
zr
et n a gi G
tl
A C
y
a z lo a
a r r a j u p l A
R A M A NI D N U C
N R E H T U O S
0 2
E
3
0
0 4
L E A A M
L A P
C O E
L O
0 5
M A C
N A S
R U T
N O C
N E C
L M E B L A
T P A
L 0 6
0 7
8
0
R A B
T
L B A V R
U A H
E 0 9
0 0 1
0 1 1
C B
K O
V
L 0 2 1
22
0 3 1
0 4 1
B
A
M 0 5 1
0 6 1
0 7 1
0 8 1
e n ot
e h
et a n o b r a
r
r
d u m fl
s
e n ni e ni m w a h S
s
s
c
a
o
e n ot
s
r fl
e n ot
s
te a n o rb a c lf e h
s
tr
ia v ul
L LI
d n a s
d u m e ni
l
lt
s
s
s
o t i r a j a P
li
e n ot
d n a
N R
st
s
s
e u q e t a u G
H T U O S
d n a s d n a
e n ot
A BB U S Y
e n ot
-
s
e n ni e ni
el a h
a lol
a m w ol l a h S
s
d n a
s
r r
s
s
s fl
e h r
s
te u O
s
et i di br u T
; i 4 e 9 t s 9 d 1 a l r i G m a f l o l e i l V a ; c s 7 e 8 9 h t 1 , o t 5 g 8 9 n 1 i d , r e o r c b c a a F e r e m t i t f a l ( a c i n o g i o t l a o r o e t g s s r t e n c e i s t e r s a p p e r s n i s l i x a a p l t a u c o i t t r h e i V w . ) 2 . m 2 k . ( g e i c F n a n t i s n i d o l i t a a t c n o o z L i . r E o n h o y i t a c d e s t n c e i h s e p r a r p s g t i t n a r e s t s e r y r p e a r i t s r i e x T a d l a n t a n s o z u i o r e o ) c h 5 a , 9 t 9 e r ) 6 1 C 9 , . 9 l : 1 a 0 ( t 1 . e 2 g g r e r O e u d p g n o o i F a C
Chapter 2
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 3 4 5 0 0 1 2 3 0 7 8 4 5 6 7 8 9 0 1 1 1 1 1 1 6 1 1 1
E N N
a c n a l b a s o R
o z a l b a T
o z a l b a T
r o b m a T s o t n a S s o L
3 1 6 1 - s a t n a f n I j a c s a C S 1 s e l a L L I H T O O F N R a r r a t i m i C E T S E W A R E L L I - o r e d a t a M D 2 R O C í p o c a Y N R E T S A E
W N O I T C E S
a s e r e T a t n a S
t l u h t f a e r a o z e a a l N a o l u p A P
Y E L L A V A N E L A D G A M R E P P U
a r m i a s m i L U
a n u L a L
i t i m i S
a c o n c a u l a b b j a a s a c P o r R A
e t l r u b a f m o r u e a l f C a e s P n a r t
a t e l l i V
A R E L L I D R O C N R E T S A E N R E H T U O S
W S S
z a P a L o a d s s a a o d l r r o l g a o u r e C M m s E
o c e s o i R e d n a u J n a S
a z o l a z r a B
a c r u M a c e S s a n a r o e y n t o e r n o H A r y F s a t i t u L a n o r r a m i C
a c i t U
o l i H s a r e h c n i r T a t e v a N
á t o c o S
t o d r a r i G
l a c O i v a Y
a s e M a z o l a z r a B
a g e t r O i n i l O . Q
a e d t n n o a H g i G a v i e N
O Z A L B N A I T - S A A N B E B L U A S D G A M
y a d n a l a u G
s a a r n e s e r t o l l A n o r l o a y F i b s H a a t i C t u L i a a l r n i r a l a u j O d u e a t p l u a A r G r e s n o M
T N E M N O R I V N E L A N O I T I S O P E D
0 0 1
A C R N I 0 A S M A A B N I B D U N S U C
t l u a o f e r e a f l a s A P n a r N C N t R I o e t a l l a u a P f r h e t f e s r a n a z r a t N
s e n o t s s d e n n a o s t s d d u n a m s d e n n a t o s s e d n u o m t s d e n n i r a s t l s a u c i v l a u d l f d n n a n a i a n l a p f l l a a t i v s a u l l o A C
R A S E M A H A B T N U I B O D U S N S U C
s e n o t s d n a s f l e h s r e n n i o t l a r o t t i L
s e t a n o b r a c f l e h s r e n n i e n i r a m w o l l a h S
1 - a n i t n e r o l F L E L M E L E Q L M E
L E M
B T
I I O I C L M C S C T C A A B H V K O C B B A L R V L A P M O O E P L E L I S M ) S U O E C A T E RC Y R A I T R E T C S A R UJ a M ( 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 7 8 e 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 g 1 1 1 1 1 1 1 1 1 A
23
m K 0 0 2
s e n o t s t l i s d n a s e n o t s d u m f l e h s r e n n i e n i r a m w o l l a h S
s e n o s t s e t d a u n m o b r d a n c a r s o e s n o e t l s a d h n s a f s l e s h t e s i d r i e b t r u u O T
e r e h t t f a o ( t g n n i o i t d a r r o o c t c a s e e r c m i t i t s a l p a s c i n i g l o a l p o e g t u o s h t t n i e s w e r ) p m e r k ( s e i x a c n l a t a s c i i t d r e l a t V . n o 2 . z 2 i r g o i h F n y a i d n o t n i t e a s c r o e L p s . t W n e s n e o r i t p . ) c e r e 5 s s 9 9 c i i x 1 h a , . p l l a a a r t g t n i e t o a r z r i e t p s r o o h y o r , C a ) i t 6 ; r 4 e 9 9 T 9 1 9 ( 1 d n g i a g l m s O a u l l o d i e n c V a a ; t n 7 e i r t e 8 C s 9 1 . d , 1 a 5 r 1 . 8 2 G 9 . f 1 : o , e r e e r u l b g a i a F c s F
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
In the Tablazo sub-basin latest Jurassic to Valanginian fluvial sedimentation was followed by mudstone deposition in marginal marine environments recording the marine transgression (Cumbre Fm, Mendoza, 1985; Renzoni, 1985c; Ritoque Fm Ballesteros and Nivia, 1985; Rolón and Carrero, 1995). Later, tidal and shallow water marine shelf carbonates (Rosablanca Fm, Cardozo and Ramirez, 1985) were deposited during Valanginian-Hauterivian time, followed by shallow marine shales (Paja Fm) during HauterivianBarremian time (c.f. Etayo-Serna et al., 1976). Although transgression was progressive from the center of the basin, two periods of relative sea retreat occurred, during Hauterivian and Aptian (Rolón and Carero, 1995; Ecopetrol et al., 1994; Figs. 2.12 t o 2.17). Later during Aptian a relative tectono-eu static sea-level rise occurred as suggested by a deeper marine facies of the upper part of the Paja Fm (Forero and Sarmiento, 1985; Ecopetrol et al., 1994; Rolón and Carrero, 1995). Berriasian to Aptian sedimentation on the Cocuy sub-basin. In the Cocuy sub-basin marine transgression started in its southern part during the end Jurassic to earliest Cretaceous as recorded by the Brechas de Buenavista Fm (Dorado, 1984) and the Calizas del Guavio Fm (Ulloa and Rodríguez, 1976; Fabre 1985a; Mojica et al., 1996). In the northern part of the Cocuy sub-basin, facies changes record the transition from continental to shallow marine sedimentation (Lutitas de Macanal Fm) during Berriasian to Valanginian times (Fabre, 1985a). During the Hauterivian to Barremian wave-dominated deltaic sandy environments of deposition developed (Arenisca de Las Juntas Fm, Fabre 1985a; Figs. 2.12 to 2.17). In Hauterivian time, the deposition of prograding sands in a rapidly subsiding basin (Fabr e, 1985a) was probably facilitated by relative tectono-eustatic base level fall. Early Cretaceous sedimentation on the Santander-Floresta palaeo-Massif. The SantanderFloresta palaeo-Massif remained emergent until Hauterivian time when deposition of continental sandstones began, followed by progradation of deltaic sandstones (Rionegro Fm, lower part of Tibasosa Fm) in turn by shallow water marine carbonates. The two sub-basins started to form a single wide basin during Hauterivian time due to flooding of the pala eo-Massif (Fabre, 1985a; Moreno, 1990a,b, 1991) and base level rise. This intrabasinal h igh was, however, a significant barrier to sediment movement until Aptian time (Cooper et al., 1995; Figs. 2.12 to 2.17). The succession of sandstone (Tambor and Los Santos Fms), limestone (Rosablanca Fm) and da rk shale (Paja Fm) facies, recorded in the Tablazo sub-basin, is laterally younger toward the east on the Santander-Floresta palaeo-Massif (sandstone: Rionegro Fm, limestone and shale: Tibú and Mercedes Fms) and in the Cocuy sub-basin (sandstone: Arenisca de Las Juntas Fm, limestone and shale: Apón Fm; Fabre, 1985a). This lateral change in age of facies occurred as a result of the oscillating and pr ogressive marine transgression toward the east during Valanginian to Aptian times (Figs. 2.12 to 2.17). Berriasian to Aptian sedimentat ion on the Cundinamarca sub-basin. Towards the south both the Tablazo and Cocuy sub-basins show a gradual increase in dark shale deposited in poorly oxygenated shallow marine environments. (Caqueza Gp, Villeta Gp; Fabre, 1985a; Rubiano, 1989; Sarmiento, 1989). In the Cundinamarca sub-basin Cretaceous sedimentation started during Tithonian?Berriasian-Valanginian with turbidite deposits in both the eastern (lower Caqueza Gp, Pimpirev et al., 1992) and western ( lower part of Utica Sandstone, Murca Fm, Sarmiento, 1989; Moreno, 1990b, 1991) flanks (Figs. 2.12 to 2.17). Turbidite deposition prevailed up to the Hauterivian in the eastern border of the basin (Caqueza Gp, Pim pirev et al ., 1992). During the earliest Cretaceous, basin subsidence exceeded sediment supply, resulting in retrogradation of the turbidite fan system, so that distal fan sediments covered middle fan mouth channel deposits. In post-Berriasian time, sediment supply increased overwhelmed basin subsidence, resulting in progradation of the turbidite fan system (Pimpirev et al., 1992) and locally by progradation of deltaic sands during Hauterivian time (upper part of Utica Sandstone, Sarmiento 1989; Moreno, 1990b). Towards the south the shallow marine sandstones and limestones of the Naveta Fm mark the development of a shoreline during Hauterivian-Barremian time (Cáceres and Etayo-Serna, 1969; Sarmiento, 1989). Differential subsidence related to syn-sedimentary normal faulting caused unstable slopes on basin margins. These processes favoured turbidite deposition during early Cretaceous up to the Aptian (lower part of Utica Sandstone, Murca Fm, Socota Fm, Pol anía and R odríguez, 1978; Sarmiento, 1989; Moreno, 1990b, 1991; Caqueza Gp, Pimpirev et al.,1 992; Figs. 2.12 to 2.17). Early Cretaceous sedimentation on the Tablazo sub-basin.
24
Chapter 2 Serrania de Perija
s a c u L n a S e d i a n r ra Bucaramanga e t l S u f a e o l a a a p r c a i n r e z m s u a a n b a S n i
1400
1300
r a il e d o r C
e
1200
1100
l r a t e n C
900
h g r i e H d n a a t t n s a e r o S l F
i s n a b b u S y u o c C l t u - f a o l e p a
Llanos Orientales Basin (LLA)
e m s t s y
Yopal
BERRIASIAN VALANGINIAN PALEOGEOGRAPHY
Positive relief (absent). Alluvial fan and fluvial sandstones and mudstones
Villavicencio
Coastal plain predominantly sandstones. Coastal plain predominantly mudstones.
-
0 a e n e d r a a c i n a a r r M e a S L
Neiva
800
800
Barinas Basin
S
Tunja
o a m a r c i a G u
N a z a Bogota r e t h p a l a e Ibague o - f a u l t
s d e n A a d i e r M
Arauca
l e l t a u d f a g n i a s e o l a M a - b a p o z b c a l a S u y a b o a B T
Manizales
1000
Cucuta
d b n u u s C
m t e s y s lt u fa o e T r a n l a s t e a r p p a a l a e o i m f a u t u l t i B
Medellin
Maracaibo Basin
900
1000
100
Littoral to inner shelf sandstones.
200 Km
Shallow marine inner shelf carbonates. Shallow marine inner shelf mudstones and siltstones. Outer shelf shales or carbonates Turbiditic
sandstones.e
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.12: Berriasian-Valanginian palaeogeography without palinspastic restoration (modified from Etayo Serna et al, 1968, Geotec, 1992; Cooper et al., 1995).
Cretaceous sedimentation started during the Aptian (Vergara and Prössl, 1994) in an extensional basin formed initially in Jurassic time. Feldspathic and lithic sandstones, conglomerates and red mudstones (Yaví Fm) were deposited as alluvial fans on valley slopes, while finer sandstones and mudstones (Alpujarra Fm) accumulated within a fluvial system flowing northward. (Florez and Carillo, 1994; Etayo-Serna, 1994; F igs. 2.16 a nd 2.17). In the whole EC basin, abrupt lateral thickness changes and ubiquitous turbidite deposition attest f or a local tectonic/differential subsidence depositional conditions in Early Cretaceous time (Figs. 2.13 to 2.17). Regional correlation of Early Cretaceous relative tectono-eustatic cycles is difficult to esta blish due to local active extensional tecto nics. Since Aptian time these relative tectono-eustatic level cycles become more tractable or traceable ( Figs. 2.7 t o 2.11). An important transgression followed a relative se a-level rise during late Aptian time. During this time the sea flooded all the area of the present EC even south of the Cundinamarca sub-basin. (Etayo-Serna et al., 1976; Etayo-Serna, 1994). During late Aptian time the sea gradually flooded the UMV and dark mudstone and limestone (El Ocal Fm) were deposited in a shallow marine environment (Etayo-Serna, 1994; Figs. 2.16 and 2.17). Dark grey to black mudstone was deposited regionally in a dysoxic shallow mar ine shelf (up per part Paja Fm, in the former Tablazo sub-basin, Fómeque Fm, in Aptian
sedimentation
on
the
palaeo-Upper
25
Magdalena
Valley
(UVM).
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
0
1400
s a c u L n a S e d ia n a r r e S
0
1300 258
ra e il l d r o C
Medellin
1200
Cucuta
Manizales
112
155 123
l t u f a e o l a a 226 p e z a r u S 698
Arauca 0
0
l t a u
Bogota Ibague
0 0 0
l a tr n e C
900
0
1000
0
0 0
300 1000
m t e y s 0 s l t 1500 a u f Yopal o e l a 0 p a 0 0 o 0 0 a m c i 0 0 a 0 u G 0
0
0
1500
0 0
0 0 0
0
0
0 0
0
0
0
0 0
0
0
0 0
0
Villavicencio 0
0
0
0
0 0 0
0
0
0
Lllanos Orientales Basin (LLA)
0 0
0
0
0
BERRIASIAN
0
0
100
200 Km VALANGINIAN
0
Neiva
0
0
0
0
0
0
1000
0
T r a n s 1110 f e r p a l a e o o f f a u l t
0
1000
S
0
0
900
258
3335
1000
Basin
Bucaramanga
798
0
Barinas
0 530 602
74
f m o te a e l s y 692 p a s a 10 lt 757 a c 440 y u o 20 fa B 600 17 o e T la Tunja 1210 r a n s a f 20 e r p p a a l a e o f im a u u l t it 2135 B 275
190
e s d A n a i d 0 e r M
0
1156
200
1100
Maracaibo Basin
Serrania de Perija
0
0
800
THICKNESS (meters)
San Jose del Guaviare 0
S
800
900
1000
1100
1200
1300
1400
1500
Figure 2.13: Berriasian-Valanginian thickness (meters) without palinspastic restoration. Thick lines represent alaeofaults believed to be active during Berriasian-Valanginian time.
the former Cocuy su b-basin, Villeta Gp, up per part Socotá Fm in the former Cundinamarca sub-basin, El Ocal Fm in UMV; Figs. 2.16 and 2.17). 3.2.2. Cr etaceous post-r ift sedimentati on
Cretaceous post-rift sedimentation is illustrated in Figures 2.7 to 2.11. Villamil (1993) interpreted limestone-shale or chert-shale rhythmic beds as Mi lankovitch cycles. Using graphical highresolution stratigraphic correlation he showed that these distal pelagic limestone-shale cycles are coeval to proximal parasequences. Assuming these cycles have all the same duration, and that subsidence was constant through time, Villamil (1993) plotted the thickness of all cycles in a modified Fisher plot (a stacking plot for cyclic rhythmic sedimentation) to obtain a curve of changes in relative accommodation space or relative tectono-eustatic base level. Based on facial analysis, macrofossil biostratigraphy, high-resolution event and cycle chronostratigraphy, together with the modified Fisher plots, Villamil (1993) proposed a sequence stratigraphic interpretation and a relative tectono-eustatic level history.
26
Chapter 2
1400
1300
r a i l e rd o C
1200
1100
Manizales
900
s a c u L n a S e d ia n a rr e S
Cucuta
Ibague
y l e l a V a n l e a d a g M r Neiva e p p U
800
900
Barinas
Bucaramanga
t l a u f a o e n a e l l p a d a g e z a i n r u a M a s S B l t o b - f - a u z u o l a S l a e b a p a T a c y a B o
Arauca
i n s m a t e B s b s y u t l S u f a y u o a e o c l C p a e S
o a m r i c a u a G
i v
r
t
Llanos Orientales Basin (LLA)
Yopal
HAUTERIVIAN BARREMIAN
N a z a r e Bogota t h p a l a e o f a u l t
PALEOGEOGRAPHY Positive relief (absent) All uvia l fan and flu vial sandstones and mudstones
Villavicencio
l r a t e n C
800
s u e q i c h g h M a T r o u
Basin
m s t e sy u lt f a a e o l T r p a a n a Tunja C u n s f u a n d e r p it S u i n a a l a e B b - m a o B a r c f a s i n a u l t
Medellin
1000
Maracaibo Basin
e h t n g a u b i r o r U T
Coastal plain predominantly sandstones Coastal plain predominantly mudstones
0
a e n e d r a a i n c a a r M r e a S L
1000
100
Litoral to inner shelf sandstones
200 Km
Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltstones Outer shelf shales or carbonates Turbiditic sandstones
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.14: Hauterivian-Barremian palaeogeography without palinspastic restoration (modified from Etayo Serna et al, 1968, Geotec, 1992 ; Cooper et al., 1995).
During Albian a relative base level fall favoured progradation of deltaic and littoral sands (Caballos Fm, Florez and Carrillo, 1994; Etayo-Serna, 1994) in the area of the UMV and the eastern border of the basin (lower part of Une Fm, Fabre, 1985a). During middle to late Albian transition from the near-shore marine facies of the Caballos Fm to the deepening upward lower part of Villeta Gp in the Upper MV recorded a rise in relative tectonoeustatic level (Villamil, 1993; Etayo-Serna, 1994). This tectono-eustatic level rise also was recorded by the upward deepening trend from the shallow water San Gil Inferior Fm t o the deeper San Gil S uperior Fm, the Socotá Fm to the Hilo Fm and within the Une Fm (Villamil, 1993 ; Figs. 2.7 t o 2.11). During late Albian-early Cenomanian a relative tectono-eustatic level fall was recor ded by progradation of the upper part of Une Fm and a generalized shallowing upward facies trend. In the earliest Cenomanian there is a sequence boundary expressed as a forced regression (unnamed shale overlying the cherts of Hiló Fm, shallow water sandstone of Churuvita Fm over deeper shale of San Gil Superior Fm, Villamil, 1993). In the Upper Cenomanian, Villamil (1993) interpreted the next marked sequence boundary (first sandstone in the shales of Villeta Gp, sandstone upper part of Churuvita Fm, uppermost sandstone Une Fm).
27
a
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Serrania de Perija
s a c u L n a S e d ia n a rr e S
1400
1300
0 1 0
260
0 0 2
Bucar amanga
782
0 0 5
283
0
900
Bogota 480
200 100
0 0 0
1300 312
Llanos Orientales 0 Basin (LLA)
0 0 2 0
1210
1040 1040
160
1246
l t u f a0
m s t e y s
0
Yopal
0
o l a e 0 a p 0
0 0 0
0
0
0
470
0
0
0
0
0
Villavicencio
0
a l n tr e C
1200
2122
227
o m r a 0 a 0 i c 1 0 a G u
318
0 0
Arauca
2 0
581 0 0
306
t l u 775 f a 627 e o l a a p 579 c a y a 315 o 613 B 265
804
Ibague
1 0
393
Tunja
Manizales 263
Basin
0
n s 0 f e r 0 p a 5 l a e o f a u l t
433
1000
Barinas
160
650 T r a
0 0 1
11
s d e n A
0 2 0 0 0
5 0 0
225
650
1100
2000
2 0 0
460 476
187
0 0 1 0 5 0 0
130
t 222 l u f a e o l a 0 p a 0 5 e z r a 612 S u
ra e i ll r d o C
Medellin
i d a e r M
2 0 0 0 5 0
304 283
0 0 1 0 0 2
1 0 0
Cucuta
284
1200
Maracaibo Basin
0 0
0 0
0
0 0
0
100
HAUTERIVIAN
0
200 Km
BARREMIAN
0 a n e
Neiva r a
800 c a M
e d 0
a
r
a L
800
THICKNESS
San Jose del Guaviare
900
r e
(meters)
i S
1000
1100
1200
1300
1400
Figure 2.15: Hauterivian-Barremian thickness (meters) without palinspastic restoration. Thick lines represent palaeo-faults believed to be active during Hauterivian-Barremian time.
During late Cenomanian, Turonian and Coniacian times the tectono-eustatic base level reached its maximum Mesozoic level. The sea flooded the entire northwestern corner of South America and dark gray shale was deposited from Venezuela to northern Peru. (Thery, 1982 in Fabre, 1985a, 1986). Villamil (1993) recognized smaller relative tectono-eustatic level cycles during this time interval. A relative tectono-eustatic base level rise during late Cenomanian (Villamil, 1993) induced a slight deepening of the basin and a notorious decrease of the detrital supply to the basin. This lead to basin starvation and the slow deposition of black laminated shale or micritic limestone pelagic facies in the distal parts of the basin. The maximum flooding surface located at the Cenomanian-Turonian boundary is characterized by a highly fossiliferous concretion horizon within the Frontera Fm and lower part of San Rafael Fm (Villamil, 1993). During Turonian-Coni acian time the present day LLA foothills were flooded (Cooper et al.,1 995) but not the entire LLA are a (Figs. 2.7 t o 2.11). From Middle Turonian to late Coniacian a gradual progradation and shallowi ng upward during deposition of the upper part of the San Rafael Fm and the Villeta Gp in the Upper MV was related to a relative tectono-eustatic level fall (Villamil, 1993).
28
Chapter 2
1400
t
l u a
1300
f
r a e il l rd o C
1200
s a c u L n a S e d a i n a r Bucaramanga r e S a e o l a p a
e t n h a g u b i r o r U T
Maracaibo Basin
s u e q i s h h g c d e u a n o M T r A a d i e r M
Cucuta
a e n l a t d l t i u g i n v f a a s o M e a l a B p a b - t o u z z e r l a S f a u l u a b S a e o l a T p a
1100
Manizales
o l a e p a o m r a á c i a G u
a z a r e t h p a l a e o Bogota f a u l t
1000 Ibague
Basin
Arauca r e S
n i s
m a s te B b s y u S l t a a u c y a f u y B o m e o c t e o y s l a Yopal s Tunja C T r l t a n p a s f f a u e a r o p a e C u l l a i m a e n d t e m o p a y s f it u S u i n a m s a u a b - B a r B l t l t a r a s i c a f a u á m N n
Medellin
Barinas
Llanos Orientales Basin (LLA)
T n a i a s C u
APT IAN PALEOGEOGRAPHY Positive relief (absent) Alluvial fan and fluvial sandstones and mudstones
Villavicencio
900
l r a t e n C
800
800
y l e l a V a n l e d a g a M r e Neiva p p U
900
0
e a d n e a r a i n c a a r r M e a S L
1000
100
Coastal plain predominantly sandstones Coastal plain predominantly mudstones Littotal to inner shelf sandstones Shallow marine inner shelf carbonates
200 Km
Shallow marine inner shelf mudstones and siltstones Outer shelf shales or carbonates
San Jose del Guaviare
1100
Turbiditic sandstones
1200
1300
1400
1500
Figure 2.16: Aptian palaeogeography without palinspastic restoration (modified from Etayo Serna et al, 1968, Geotec, 1992; Cooper et al., 1995).
In the Upper MV during the late Coniacian to Santonian the transition from the uppermost Villeta Gp, deposited in an inner shelf, to the lower chert unit of the Olini Gp, deposited in a deeper middle shelf (Jaramillo and Yepez, 1994; Ramirez and Ramirez, 1994), points to a deepening of the basin and relative tectono-eustatic level rise. ( c.f.E tayo-Serna, 1994, his Figure 2). During the Santonian, Campanian, Maastrichtian and Paleocene a general regression and progradation was recorded by littoral to transitional coastal plain facies (Guadalupe Gp, Guaduas Fm). The Guadalupe Gp sands represent two cycles of westward shoreline progradation, aggradation and retrogradation, dominated by high e nergy quartz-rich shoreface sandstones derived from the Guyana Shield (Cooper et al., 199 5; Figs. 2.7 to 2.11). Regression did not occurred continuously but with minor transgressive events recorded by fine-grained si liceous and phosphat ic facies (Föllmi et al., 1992; Plaeners Fm, Olini Gp, upper part of La Luna Fm; Figs. 2.7 to 2.11). A sequence boundary occurs at the base of the medium shale unit (Lower-Middle Santonian according to Villamil, 1993 or late Santonian-early Campanian according to palinostratigraphy by Jaramillo and Yepez 1994; Etayo-Serna, 1994) of the Olini Gp and the shallow water El Cobre sandstones of Barrio and Coffield (1992) in the UMV (Villamil, 1993). The shallow water marine sands of the Arenisca Dura Fm represent a lower forced regression system tract (sensu Posamentier et 992; Cooper et al., 199 5; Figs. 2.7 to 2.11). al.,1
29
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Serrania de Perija
s a c u L n a S e d ia n a rr e S
1400
1300 227
1200
2
Medellin
227
0
1100
9
0 7 0 1575
7 7 4
175 176
Ibague
140
3
Arauca
0 0 7
l a e p a a a r 0 m a T 0 n a a i 0 s 0 C u 0
458
0
0
4
0
0 0 0
Llanos Orientales Basin (LLA)
Yopal
0 0 0 0
0 4 0 0
240
0 200
0
0 2 0
575 73
0
0
Villavicencio 0
386
0 215
0
195
121
Basin
0
45
Bogota
0
e a d n e a r a i n c a a r r M e a S L
800 0
900
0
0
0 1 0 0
Neiv a
800
0 Bucaramanga 0 2
476
Barinas
0 0 1
220 211
154
5 0 0
208
900
185
0 2 0 0 5 0
110
214
l r a t e n C 0
0 1 0
0 0 1
0 T r a 0 0 0 4 n s 5 f e r p a l a e o f a u l t
840
Manizales
1000
s d e n A 0 4 0 a 0 r i d 2 0 e M
Cucuta
m 212 217 s t e 831 s y 930 190 0 u lt 4 0 349 f a 465 0 0 486 eo 840 1 2 0 0 0 0 0 a la 0 4 0 0 1 310 2 p 113 0 o 306 5 0 m 714 16 m 180 ra t e 0 7 0 y s 182 i ca 140 s 640 Tunja 0 u a a u l t G o f 2 0 410
r a e i ll r d o C
0 0 0 1 0
Maracaibo Basin
1000
100
AP TIAN
200 Km
THICKNESS
(meters)
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.17: Aptian thickness (meters) without palinspastic restoration. Thick lines represent palaeo-faults relieved to be active during Aptian time.
Mudstones of the upper part of the Arenisca Dura Fm and shales of the Plaeners Fm represent a transgressive system tract (Cooper et al., 1995). During Santonian-early Campanian a maximum flooding surface and a relative tectono-eustatic level rise from the medium shale unit of the Olini Gp to the Upper Chert unit has been interpreted by Villamil (1993). In contrast to the EC, where the Cretaceous maximum flooding surface occurred at the Cenomanian-Turonian boundary, the maximum flooding in the LLA occurred during the Campanian (CS at the top of Gachetá Fm, Fajardo et al.,1 993; Cooper et al.,1 995 their Fig. 4). During late Campanian the relative sea-level continued to drop and shallow marine oxygenated environments prevailed in the EC. The Labor Fm represents a sand dominated forced regression system tract (Cooper et al.,1 995). The regional regression and long term relative tectono-eustatic level fall was interrupted by a small cycle of relative tectono-eustatic base level rise during late Campanian or early Maastrichtian (?) as suggested by Föllmi et al.( 1992) and Villamil (1993) during deposition of the “Upper Plaeners Fm”. According to Cooper et al . (1995) the Upper Plaeners unit represents a condensed marine mudstone deposited during a relative tectono-eustatic level rise. During early (?) Maastrichtian time the eastern part of the basin was filled by the littoral quartz sands of the Arenisca Tierna Fm (Fabre, 1985a). According to Cooper et al. (1 995) the latter represents the transgressive systems tract of the next sequence ( Figs. 2.7 t o 2.11). The gradual uplift of the
30
Chapter 2
western margin of the UMV supplied clasts of metamorphic rocks that were accumulated by fluvial systems close to the sea in a braided delta (Cimarrona Fm, Gómez and Pedraza, 1994). Sands were dispersed along a littoral belt (Monserate Fm, Ramírez and Ramírez, 1994), while in the more distal areas carbonate silt (Díaz, 1994a) or mud (Umir Fm) accumulated (Etayo-Serna, 1994).
4. SUBSIDENCE ANALYSIS The stratigraphic record provides information about vertical crustal movements and/or sea-level changes within a basin. Basin subsidence or basement subsidence is the result of both a thermomechanical component called tectonic subsidence and a component due to sediment and water loading. Tectonic subsidence is the undistorted basin subsidence that would have occurred in the absence of sedimentation, therefore it is related to the geodynamics of the basin. In order to quantify the tectonic component of subsidence of the studied basin the 1D backstripping technique was used (Steckler and Watts, 1978; Bond and Kominz, 1984). The method is explained in Sclater and Christie (1980), Bond and Kominz (1984) and Bessis (1986). For this purpose, tectonic subsidence has been calculated from the stratigraphic record adopting local isostasy to correct for the effect of sediment loading. Corrections for compaction were made using porosity-depth relationships on the basis of the observed lithologies using standard mean exponential relations and material parameters (cf. Sclater and Christie, 1980). Most of stratigraphic columns are from published literature; well data are from Ecopetrol. As pointed out by Restrepo-Pace and Villamil (1998), and locally demonstrated by Restrepo-Pace (1989), thickness measured from outcrops in the EC can be exaggerated due to structural thickening. Cretaceous thicknesses are not well controlled (Fabre, 1986, 1987; Restrepo- Pace, 1989). In order to eliminate or reduce that possible source of error I carefully checked thickness of each stratigraphic column with available geological maps to avoid structural repetitions. I also checked the consistency of thickness between neighbouring sections. Villamil (1993) described in detail four partial (Albian to Santonian) stratigraphic columns. I checked thickness of stratigraphic columns reported in the literature with these well-constrained sections. In the four sections described by Villamil (1993), only the Cocuy stratigraphic column is located in a depocenter. The other sections studied by Villamil (1993) are close to a palaeo-high (Villa de Leiva section) or in the less subsiding southern border of the basin (Mesitas, Olini and Yaguará sections). Based on the assumption of a Late Cretaceous passive margin, Villamil (1993) estimated clinoform profile geometry during Late Cretaceous with a slope of 0.057 °. This value indicates an almost horizontal palaeo-sediment surface profile. Thus the difference in thickness without considering this slope is very small. The effects of palaeo-bathymetry have been taken into account, using sedimentary facies and faunal content as interpreted in literature. Detailed sedimentological analysis of the deepest facies concluded that palaeo-water depths never reached values greater than 200 m (Sarmiento, 1989; Villamil, 1994). Most of the Triassic and Jurassic sedimentary record is continental, and the marine facies are mostly Cretaceous, with prevailing water depths 0-75 m (Sarmiento, 1989; Villamil, 1994). Errors in water depths are likely to be less than 50 m. Sea-level changes were estimated from the curve proposed by Villamil and Arango (1998) for the Cretaceous of Colombia and Haq et al. (1987) for other time intervals. Ages are based on the data given in literature mainly on the regional stratigraphic cross-sections presented by Cooper at al. (1995). In order to express ages in Ma I used the geological time scale proposed by Gradstein and Ogg (1996). Unconformities are also included with their age and duration in Ma. Additional parameters for forward modelling are initial crust and lithosphere thickness and densities. Table 2.2 sh ows additional parameters used in forward modelling, which are average accepted values for normal continental lithosphere. Pulses of fast tectonic subsidence of the basement have been interpreted in terms of tectonic or eustatic processes. Only those pulses of fast tectonic subsidence correlatable to normal fault activity were interpreted as produced by lithosphere extension and the slower subsidence, generally later, was interpreted as produced by thermal re-equilibration of the lithosphere following the thermal anomaly created by stretching.
31
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
MODEL PARAMETERS
VALUE
Initial lithospheric thickness
120 km
Initial crustal thickness
35 km
Asthenospheric temperature
1333 °C
Thermal diffusivity
1 × 10 -6 m2 s-1
Surface crustal density
2800 kg m-3
Surface mantle density
3400 kg m-3
Sea water density
1030 kg m-3
Thermal expansion coefficient
3.2 × 10 -5 °C
Table 2.2. Parameters used to calculate tectonic subsidence in the forward model.
Tectonic subsidence refers to the vertical downward movement of the basement top surface underlying the sedimentary fill of the basin accumulated during the time interval being studied. As the studied time interval is Mesozoic, for modelling purposes I considered tectonic subsidence referred to the top surface of any pre-Mesozoic rock. Pre-Mesozoic rocks crop out only in local areas of the EC (Santander-Flor esta and Quetame-Garzón Massifs and the core of some anticlines in the flanks of the EC) (Fig. 2.1). Although the geodynamic igneous/metamorphic basement is not such pre-Mesozoic top sur face due to the local presence of highly compacted Palaeozoic sedimentary rocks, I did not include this Palaeozoic record due to large uncertainty in their age, thickness, facies and lateral continuity. This implies to neglect any compaction of these rocks during Mesozoic and Cenozoic time, which probably was very small. Since most compaction occurs early just after deposition, most compaction of these rocks occurred during Palaeozoic time. I extended downward the Cretaceous stratigraphic columns down to the basement, interpolating data from the nearest stratigraphic columns and using the palaeogeographic and restored thickness maps. For areas absent of surface outcrops, I extended the most representative Cretaceous columns down to the pre-Mesozoic top surface according to the palaeogeographic and isopach reconstructions. For these stratigraphic columns I studied the tectonic subsidence referred to a pre-Mesozoic “basement”. Additionally, to constrain the tectonic subsidence evolution during the Cretaceous avoiding uncertainties reconstructing Triassic and Jurassic section, I studied the Cretaceous tectonic subsidenc e referred to the top surface of any pre-Cretaceous rock. In the Cundinamarca sub-basin (Figs. 2.1, 2.12 and 2.18) a relatively complete Cretaceous stratigraphic section crops out in the east ern and western flank s of the mountain range, yet in the axial region the deepest wells have not penetrated the complete Cretaceous section. Stratigraphic information for the missing section in the axial area has been extrapolated from the exposures at the flanks of the EC. Relative sea-level and palaeo-water depth effects. One common difficulty with subsidence analysis is the need to separate the effect due to eustasy from tectonics ( e.g. Parkinson and Summerhayes, 1985). In order to investigate the effects of sea-level changes I used the best available data from selected stratigraphic column s (Table 2.3) to calculate the tectonic subsidence both with and without the incorporation of sea-level chan ges (Fig. 2.19). Comparison of the tectonic subsidence curves with the palaeo-water depth curves and also with sea-level curves (Fig. 2.19) suggests the f ollowing: (1) In gener al the amplitude of the tectonic subsidence curve is greater than that of the sea-level curve (e.g. Cocuy section, Fig. 2.19). Only where tectonic subsidence is small, the magnitude of sea-level changes is of the same or der as short wave tectonic subsidence variations ( e.g. Q Olini section, Fig. 2.19). Thus calculated tectonic subsidence in stratigraphic columns with small tectonic subsidence is more sensitive to sea-level changes. For those sections with relatively minor tectonic subsidence (comparable in magnitude with sea-level changes), a sea-level correction is necessary in order to separate the tectonic and sea-level signals.
32
Chapter 2 Serrania de Perija
s ca u L n a S e d ia n ra r e S
1400
1300
Maracaibo Basin
11
e s d A n i d a e r M
600- Cucuta 800
7
12
600-
2 1000
Bucaramanga Arauca
1 100-
Llanos Orientales Basin (LLA)
500
9 1200-
1200
Barinas Basin
Medellin
3
r a ll e i r d o C
1800
8001700
10
A
100200
Tunja
1100 Yopal
4 11001800
1000
5 750-
Ibague
Bogota 600-
B
8 1400
1200
6
300850 Villavicencio
900
l r a t e n C 1 0 100200
Basin compartment
Neiva
800
800
900
200Km
10 0
0
Manizales
1000
Cretaceous tectonic subsidence (m)
1100
BASIN COMPARTMENTS 1. CENTRAL CORDILLERA 2. MIDDLE MAGDALENA VALLEY 3. MAGDLENA-TABLAZO SUB-BASIN 4. CUNDINAMARCA SUB-BASIN 5. SOUTHERN CUNDINAMARCA SUB-BASIN 6. UPPER MAGDALENA VALLEY 7. SANTANDER FLORESTA MASSIF 8. S CUNDINAMARCA AND W COCUY SUB-BASINS 9. COCUY SUB-BASIN 10. LLANOS ORIENTALES 11. MARACAIBO (CATATUMBO) SUB-BASIN 12. MACHIQUES TROUGH
1200
1300
1400
1500
Figuere 2.18: Cretaceous basin compartments (sub-basins) and their tectonic subsidence in meters. Section A shows the tectonic subsidence of the Tablazo and Cocuy sub-basins. Section B illustrates the tectonic subsidence of the Cundinamarca sub-basins.
(2)
(3)
Uncertainty in palaeo-water depth increases with palaeo-water depth (as accuracy in palaeo bathymetry estimation decreases with increasing water depth). However, in the cases here analysed the absence of truly deep palaeo-depositional environments makes this uncertainty small. Comparing the tectonic subsidence curves calculated involving sea-level changes reveals that in many cases differences in the subsidence curves attributable to the latter are beyond the time resolution of the database. The frequency of the sea-level changes proposed by Villamil and Arango (1998) for the Cretaceous of Colombia and by Haq et al. (1987) are in general higher than those changes inferred in subsidence, due to the smaller number of data points in the subsidence curves. Clear identification of a high frequency eustatic signal would require more accurate time resolution in the data and adds very little information regarding regional crustal studies. Taking into account the resolution of the data, tectonic events should be limited to those clearly different from the sea-level signal. Pulses of tectonic subsidence with amplitude and frequency of the same order of magnitude as the sea-level variation curve are difficult to identify without filtering the sea-level curve signal.
33
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
In order to distinguish small tectonic events, a sea-level correction would be necessary. However a correction becomes impractical for the following reasons: (1) Less time resolution in most of the stratigraphic columns compared to the sea-level curve; (2) Controversy about the precise significance of the seal-level cycles (Miall, 1992, 1993); (3) Differences between the global sea-level curves (Haq et al., 1987) and the cycles proposed by Villamil and Arango (1998) for the Cretaceous of Colombia, which can include regional tectonic effects; and (4) Uncertainty in the magnitude of sea-level variations in the sea-level curves. For practical purposes, therefore, in order to compare tectonic subsidence patterns of most of the stratigraphic columns, the general analysis was done without including detailed sea-level effects. It is assumed that the identified stretching events mainly represent a tectonic signal because: (1) (2)
Each of them occurs in specific tectonic active portions of the basin and is not uniformly and synchronically over the whole area as would be expected for eustatic changes; and Each of them can be correlated in space and time with normal fault activity.
Late Cretaceous eustatic sea-level rise may have contributed to the observed basin-wide subsidence (Fabre, 1987), but this could also be the result of intraplate compressional stress (Cloetingh et al ., 1985). In addition short-term sea-level fluctuations may have played an important role in the sedimentation distribution. The occurrence of such short-term and relatively low magnitude (compared to tectonic effects) sea-level fluctuations can not be easily detected from the subsidence data. CRETACEOUS SUB-BASINS 1. Central Cordillera 2. 3. 4. 5.
Middle Magdalena Valley Magdalena-Tablazo Cundinamarca Southern part Cundinamarca
6. Upper Magdalena Valley 7. Santander-Floresta Massif 8. Eastern Cundinamarca and Western Cocuy 9. Cocuy 10. Llanos Orientales
STRATIGRAPHIC COLUMN
REFERENCES
San Félix
Rodríguez and Rojas (1985), Etayo-Serna (1985) Tablazo Ward et al.( 1973), Ecopetrol et al. (1994) Villa de Leiva Etayo-Serna (1968), Villamíl (1994) Villeta/Yacopí Sarmiento (1989), Rubiano (1989) Apulo Cáceres and Etayo-Serna (1969), Villamíl (1994) Q Olini Villamíl (1994), Etayo-Serna (1994) Tibasosa Renzoni (1967), Renzoni et al. (1967) Cáqueza/Paz de Río Hubach (1957), Renzoni (1965a,b) Cocuy/Guateque Arauca-1
Fabre (1986), Ulloa and Rodríguez (1976), Villamil (1994) Ecopetrol and Beicip (1995)
Table 2.3. List of stratigraphic columns used to study sea-level effects .
4.1. TECTONIC SUBSIDENCE DURING TRIASSIC AND JURASSIC TIME For the subsidence analysis of the Triassic sedimentary record I assumed, following Geyer (1982), that the poorly fossiliferous sediments (El Sudán Fm.) of the Ciénaga de Morrocoyal north of the Serranía de San Lucas, with lithology and relative stratigraphic position similar to those of the Luisa Fm. of the Payandé region, are Triassic and time-correlative. I also assumed that Triassic sediments were accumulated in the western flank of the EC. With these hypotheses is easier to explain the Jurassic and Cretaceous subsidence in the western part of the EC, as partially produced by thermal subsidence of an earlier rifting event. Further discussion is based on this assumption. The tectonic subsidence curves obtained for Triassic and Jurassic times are shown in Figure 2.20.
34
)
Chapter 2 (
APUL O Air loaded tectonic subsidence
APULO
Age(Ma) 14 0
12 0
10 0
80
60
40
20
-150
0
h
0
-100
10 0 20 0
16 0
p
30 0 m
14 0
12 0
10 0
80
60
40
20
0
-50
e 40 0 t
0
50 0 d
50
60 0
10 0
70 0 ) 15 0
80 0 90 0 1000
20 0 (
Q OLINI Air lo aded tectonic subsidence
Q OLINI
Age(Ma) 14 0
12 0
10 0
80
60
40
20
-150
0
0 h
-100
10 0
16 0
14 0
12 0
10 0
80
60
40
20
0
20 0 p
-50 0
m
e 30 0 t d
50
40 0
10 0 )
15 0
50 0
20 0
60 0 (
TIBAS OSA A ir loaded tectonic subsid ence Age(Ma) 14 0
12 0
10 0
80
40
60
20
TIBASOSA -100
0
0 10 0
-50
20 0 30 0 p 40 0 m
16 0
14 0
12 0
10 0
80
60
40
20
0 0
e 50 0 t
50
60 0 d 70 0 80 0
10 0
90 0 1000
15 0
P A Z D E R I O A ir l o a d e d t e c t on i c s u b s i d en c e
PAZ DE R IO
Age(Ma) 16 0
14 0
12 0
10 0
80
60
40
20
0
-150 0 -100
20 0 40 0
-50
p
60 0 m et
16 0
14 0
12 0
10 0
80
60
40
20
0 0
80 0 d 1000
50
1200 10 0
1400
Figure 2.19: Effect of sea-level changes and palaeo-water depth on tectonic subsidence curves. On the left tectonic subsidence curves. Left panel: dashed red line: tectonic subsidence including sea-level and palaeo-water depth. Dotted blue lines: tectonic subsidence without considering either sea-level or palaeo-water depth. Continuous black line: tectonic subsidence considering palaeo-water depth but not sea-level change. Right panel: blue line: sea-level curve (modified after Haq et al., 1987 and Villamil and Arango, 1998) and black line: palaeo-water depth.
35
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes e p
CAQUEZA
C A Q U E Z A A i r l o a d e d t e c t o n i c s u b s i d e nc e t
-100
age(Ma) 160
14 0
12 0
10 0
80
60
40
20
-50 h
0
16 0
14 0
12 0
10 0
80
60
40
20
0
0
0 50 0
50
1000 d
10 0
1500 e
2000
15 0 p
COCUY
COC UY Air loaded tectonic subsidence age(Ma)
-150 t
16 0
14 0
12 0
10 0
80
60
40
20
0
0
-100
50 0
-50 16 0
14 0
12 0
10 0
80
60
40
20
0
1000
0 d
1500
50
2000 e
10 0
2500
15 0 p
G U A T E Q U E A i r lo a d e d t e c t o n i c s u bs i d e n c e
GUATEQUE
age(Ma) 16 0
14 0
12 0
10 0
80
-100 60
40
20
0
0
-50
20 0
16 0
14 0
12 0
10 0
80
60
40
20
0
40 0
0
60 0 80 0
50 d
1000 1200
10 0
1400 1600 e
1800
15 0
A R A UC A - 1
AR AUC A- 1 Air loaded tectonic su bsidence
-100
age(Ma) 10 0
80
60
40
20
0
0
-50
20 0
16 0
14 0
40 0
12 0
10 0
80
60
40
20
0 0
60 0 80 0 1000
50 d
10 0
1200 1400
15 0
1600
Figure 2.19: Continued
36
Chapter 2 4.1.1. Basin compar tments
Similarities in the shape of the subsidence curves related to their geographic position confirm that Triassic-Jurassic sedimentation occurred in two separate basin compartments, each one with its own subsidence history and sedimentary fill: (1) Upper Magdalena and Cienaga de Morrocoyal (region A in Fig. 2.20). Subsidence curves of the Upper Magdalena-Cienaga de Morrocoyal show t wo fast subsidence events (Fig. 2.20). The total tectonic subsidence that oc curred in this area dur ing Triassic-Jurassic time varies between 100 m up to a maximum of 1000 m (Fig. 2.20). (2) East ern Cordillera (Regio n B in Fig. 2.20). Su bsidence curves suggest three fast subsidence events (Fig. 2.20). Total te ctonic subsidence that occurred during Triassic-Jurassic time in this region varies from 100 m to 1500 m (Fig. 2.20). 4.1.2. Tr iassic-Ju r assic subsidence events
Distinct tectonic subsidence phases in different basin compartments demonstrate the dominant role of tectonic processes on subsidence. In this section I identify fast subsidence events and compare them to the regional geological observations to see if they can be interpreted as produced by lithosphere extension. Because of the limited number of control points defining the subsidence curves it is not possible to define correlative events of increased subsidence in all the curves. However, most curves suggest that at least three events of increased subsidence occurred in various sectors of the study area (Fig. 2.20). A simpler interpretation for these curves would be a single increased subsiding event from the Triassic u p to Middle Jurassic, for a time span of more than 70 Ma. Based on rapid lateral changes of thickness and facies several authors (Macía et al., 1985; Bayona et al., 1994; Mojica et al., 1996, etc. ) have proposed that the Triassic-Jurassic sediments accumulated in narrow rift basins. Guillande (1988) described from outcrops in the UMV Jurassic normal faults affecting the Saldaña Fm. Drastic differences in the subsidence curves for the different columns clearly indicate fault control on subsidence. Rapid subsidence in rift basins usually is the result of lithosphere stretching ( e.g.Z iegler, 1994). Thus the subsidence events were produced by lithosphere stretching phases. Each stretching phase affected more intensively specific parts of the basin as shown by the subsidence curves. However, there are unconformities separating several stratigraphic intervals deposited during Triassic and Jurassic times as indicated by Cooper et al. (1995, their Figs. 4 and 6). These unconformities may represent periods of reduced su bsidence. Tables 2.4a a nd 2.4b s how the lithospheric stretching events interpreted from the subsidence curves. The lithospher e stretching events are (Fig. 2.20): Triassic event (although variable in different columns, comprised between ∼248 to ∼235 Ma, time scale of Gradstein and Ogg, 1996, see data limitations in section 7.3). This event is best represented in the Upper MV-Cienaga de Morrocoyal. Subsidence curves ( Fig. 2.20), thickness variations and the location of the present day fault patterns ( Fig. 2.28) suggest that small narrow rift basins formed on opposite sides of the present-day Magdalena-La Salina fault system, which probably acted as normal master faults (Fig. 2.28). Obviously a complete network of normal faults is of common occurrence in rift systems. Abrupt lateral changes of sediment thickness in the Upper Magdalena Valley suggest differential subsidence at different faulted blocks (Mojica et al., 1996). The relative location of these basins on opposite sides of major fault systems suggests that they probably were separated by accommodation transfer zones ( e.g. accommodation zone between the Rufizi and Kivu grabens in the Western Rift arm of East African Rift system, Ebinger, 1989). Most rift basins consist of half graben geometry, the polarity of which often changes along trend across accommodation transfer zones (Ziegler, 1994). If the relative location of these rift basins has been preserved, their present map distribution would suggest an en-échelon pattern. The hypothesis of a depocenter in the Cienaga de Morrocoyal area is based on the assumption that early Triassic sedimentary rocks occur in this area as suggested by Geyer (1982). The only Early-Middle Triassic sedimentary record, in the area of this study, is the UMV Luisa Fm rocks according to Mojica et al.( 1996).
37
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes e A . U PPER M A GD A LENA VA L LE Y M O R R O C O Y AL ( W E S T E R N PA R T OF C H I BC H A TE R R A NE) Ag e (M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
2 San Luc as
16 0
15 0
14 0
6 Alpujarra
13 0
12 0
6 Prad o 6 Coello Luisa
0
6 Neiva
6 Atac o
50 0 6 Payande 6 Q Olini
2 E l S u d a n 6 R Sal daña M o r r o c o ya l
1000
1500
T
B. WESTERN EASTERN C ORD ILLERA (EASTERN P ART OF CHI BCH A TERR ANE) Ag e (M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
2 C a s c a j a l e s
15 0
14 0 13 0 12 0 3 Chima 8 Buenavis ta 3 Onzaga 2 San Lucas 9 Bata 3 Medios 3 La Rus ia
0
50 0
1000
2 Cimitarra
1500
T
Figure 2.20: Tectonic subsidence curves obtained from the Triassic-Jurassic sedimentary record. Horizontal axis represents age in Ma. Vertical axis represents tectonic subsidence in meters obtained from backstripping analysis. Note also the vertical bars representing the fast tectonic subsidence events. Horizontal segments in the subsidence curves represent times of no deposition without tectonic subsidence. A. Upper Magdalena Valley-Morrocoyal (western part of Chibcha Terrane of Toussaint, 1995, see Fig. 2.5). B. Western flank Eastern Cordillera (eastern part of Chibcha Terrane of Toussaint, 1995, see Fig. 2.5). Location of stratigraphic columns with Triassic and Jurassic section is shown on Figures 2.2 and 2.5.
Acceptance of this idea would imply an alternative hypothesis of a single narrow rift basin located in the area of the UMV as suggested by the latter authors. Latest Trias sic to Middle Jurassic event (comprised between ∼208 – ∼185 Ma). Subsidence curves (Fig. 2.20) stratigraphic thickness and fault distributio n (Fig. 2.29) suggest that narrow isolated rift basins were located on opposite sides of the Magdale na-La Salina fault system. If the relative location of these rift basins would be preserved, their pre sent map distribution would suggest an en-suggest an en-échelon pattern. The width of these basins increased as compared to early Triassic ones (Mojica et al., 1996). A fast subsidence favoured marine ingression. Volcanic rocks were abundant in this period. Middle Jurassic event (approximately ∼180 – ∼176 Ma). Palaeogeographic and stratigraphic thickness distribution indicates continued widening of rift basins. However they still remained relatively narrow. Major depocenters developed in the MV and western flank of EC probably existed within elongated NNE rifts on opposite sides of the Magdalena-La Salina fault system ( Figs. 2.20 and 2.30). Major depocenters also developed in the palaeo-MV and the current location of t he western flank of EC probably in elongated NNE rifts. The palaeo-La Salina, palaeo-Suarez and palaeo-Boyaca fault systems possibly limited those rift basins. According to some authors, small isolated rift basins developed at other locations in the study area: e.g. small grabens in the area of the Santander Massif, (Boinet, 1985; Boinet at al . 1985; Geotec, 1992; Kammer, 1993b), Perijá (Shagam, 1975; Maze, 1984), Mérida Andes (Ricardi et al., 1990, in Mojica et al., 1996), LLA (Numpaque, 1986 in Cooper et al., 1995; Geotec 1992) and Maracaibo (Shubert and Ricardi 1980, in Mojica et al., 1996) areas. Volcanic activity decreased at this time, mainly occurring in the Mérida Andes (basalts in La Quinta Fm, Maze, 1984).
38
Chapter 2
Basin compartments Sub-basins * A A A A A A A A A B B B B B B B B B
1 6 6 6 6 6 6 6 6 1 2 3 3 3 3 7 8 9
Event 1
Stratigraphic column
248.2
235
Stretching factor â ä 1.23 1.23
240 248.2 240
235 235 235
1.091 1.091 1.061 1.061 1.075 1.075
Start End (Ma) (Ma)
El Sudán Alpujarra Ataco Coello-Luisa Neiva Prado Q El Cobre Q Olini R Saldaña San Lucas Cascajales-1 Arcabuco Chima Medios Tablazo-Lebrija Riolitas Onzaga Cáqueza Santa Maria de Batá
Event 2
208 208
207 185
Stretching Start End factor (Ma) (Ma) â ä 1.088 1.088 1.045 1.045
208
185
1.064 1.064
208 208 208
207 1.042 1.042 207 1.122 1.122 185 1.061 1.061
Start End (Ma) (Ma)
248.2 248.2 248.2
235 235 235
1.165 1.165 1.168 1.168 1.163 1.163
248.2 238
235 235
1.132 1.132 1.046 1.046
238 248.2
235 235
1.044 1.044 1.114 1.114
Event 3
208 210
185 185
Stretching factor â ä
177 180.1 159.4 180.1 177
176 179 149 179 176
1.349 1.09 1.008 1.072 1.387
1.349 1.09 1.008 1.072 1.387
150
149
1.023 1.023
1.025 1.025 1.111 1.111
* Triassic-Jurassic sub-basins: A Upper Magdalena and Cienaga de Morrocoyal (western part of Chibcha Terrane); B Eastern Cordillera (eastern part of Chibcha Terrane). Numbers refer to Cretaceous sub-basins shown in Fig.2.18 Table 2.4a Mesozoic stretching events and stretching factors. Triassic and Jurassic stretching events and lithosphere stretching factors calculated from stratigraphic columns where the entire Mesozoic sedimentary record is inferred to be present.
Basin compart sub-basin # 1 1 1 2 2 " 2 2 2 2
Stratigraphic column Abejorral San Felix San Pablo Casabe-199 Cascajales-1 " " " Infantas-1613 Lebrija-1 Llanito-1 Sabalo-1K
Event 4 Stretching Start End factor (Ma) (Ma) â ä 144 132 2.697 1.237 141 132 1.024 1.024 142 132 1.081 1.081 144 127.8 1.522 1.146 144 127.8 1.059 1.113 * 144 127.5 0.987 0.987 138 127.8 1.301 1.117 144 127 1.625 1.213 138 127.8 1.313 1.098 144 127.8 1.37 1.113
Event 5 Stretching Start End factor (Ma) (Ma) â ä
Event 6 Start (Ma)
End (Ma)
Stretching factor â ä
112.2 104.4 1.004 1.004 114 109.3 114 109.3 114 109.3 114 109.3 114 109.3 114 109.3 114 109.3
1.03 1.607 1.132 1.328 1 1.478 1.039
1.03 1.205 1.094 1.113 1 1.131 1.039
# Numbers indicate Cretaceous sub-basins as shown in Fig. 2.18. * Modelled using Triassic and Jurassic actual or infer red stratigraphy. Table 2.4b Mesozoic stretching events and stretching factors. Cretaceous stretching events and stretching lithosphere factors from stratigraphic columns where the Cretaceous sedimentary record is present.
39
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Basin compart sub-basin # 3 " 3 " 3 3 " 3 3 3 " 3 3 3 4 4 4 " 4 4 " 4 4 4 4 4 " 4 " 5 5 6 " 6 " 6 6 6 6 6 " 6 6 " 6 6 " 6 " 7 7 7 7 7 7
Stratigraphic column Arcabuco " " " Chima " " " Cimitarrra Los Medios " " " Los Santos Simacota Tablazo " " " Vadorreal Velez Villa de Leiva Chitasuga-1 La Calera Quipile " " " Simijaca Suba-2 " " " Suesca-1 Suesca Norte-1 Sutamarchan Tabio Villeta " " " Yacopi " " " Apulo Fusagasuga Alpujarra " " " Coello " " " Girardot Guataqui Itaibe Melgar Neiva " " " Ortega Prado " " " Q Calambe Q El Cobre " " " Q Olini " " " Chivata Cormichoque-1 Floresta Guaca Matanza Tibasosa
Start (Ma) 140 * 140 144 * 144 143 142 * 142 136.5 144 144 * 144 134.5 142 142 142 142 141 * 141 142 142 * 142 142 142 142 142 141 * 141 134.2 * 134.2 127.6 127.6
Event 4 Stretching End factor (Ma) â ä 127.5 1.209 1.209 127.5 1.099 1.099 127.5 1.43 1.265 127.5 1.47 1.261 127.5 1.354 1.31 127.5 1.111 1.111 127.5 1.043 1.043 127.5 1.131 1.088 127.5 1.251 1.251 127.5 2.702 1.348 127.5 1.052 1.052 127.5 1.585 1.179 127.5 1.151 1.245 127.5 2.016 1.332 130 1.196 1.196 130 1.331 1.206 130 1.298 1.126 130 1.511 1.147 130 1.247 1.247 130 1.29 1.194 130 1.085 1.085 130 1.217 1.217 130 1.214 1.214 130 1.234 1.234 130 1.227 1.227 127.5 1.853 1.226 127.5 1.01 1.01 127.5 1.665 1.665 127.5 1.449 1.449 127 1.792 1.164 127 1.571 1.13
* *
* * * *
Event 5 Stretching Start End factor (Ma) (Ma) â ä 114.8 112.2 1.069 1.011 114.8 114.8 114.8 114.8 114.8 114.8 114.8 114.8 114.8 114.8 114.8 114.8 121 121 121 121 121 121 121 121 121 121 121 121 121 121 121 121 121 115.1 115.1 121 121 121 121 108.5 121 119.1 121 121 121 121 119.1 119.1 121 121 121
112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 112.2 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6 102.6
132.8 127 3.144 1.215 122.4 122.3 2.029 1.179 132.6 127 1.024 1.037 132 127 1.259 1.075 128.8 127 1.716 1.13 132.8 127 2.046 1.181
# Numbers indicate Cretaceous sub-basins as shown i n Fig. 2.18. * Modelled using Triassic and Jurassic actual or inferred stratigraphy. Table 2.4 b.: Continued
40
1.082 1.063 1.329 1.256 1.096 1.009 1.064 1 1 1.374 1.219 1.052 1.536 1.58 2.046 1.301 3.238 1.622 1.221 2.476 1.824 1.825 1.587 3.575 1.52 10 1.77 1.261 1.466 1.37 1.411 1.122 1.11 1.366 1.366 1.176 1.18 1.407 1.312 1.658 1.513 1.422 1.477 1.154 1.275 1.137 1.166
1.026 1.019 1.111 1.077 1.043 1.009 1.01 1 1 1.08 1.072 1.052 1.187 1.178 1.281 1.115 1.318 1.187 1.142 1.304 1.224 1.252 1.206 1.64 1.492 1.406 1.299 1.049 1.071 1.243 1.222 1.113 1.066 1.157 1.173 1.087 1.121 1.124 1.082 1.284 1.145 1.115 1.1 1.093 1.024 1.128 1.049
Event 6 Start (Ma)
End (Ma)
Stretching factor â ä
Chapter 2 Basin compart sub-basin # 7 7 8 8 " 8 8 8 9 9 9 9 9 9 9 9 9 9 " 9 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
Stratigraphic column
Start (Ma)
Event 4 Event 5 Stretching Stretching End Start End factor factor (Ma) (Ma) (Ma) â ä â ä 127 2.204 1.185 127 2.576 1.213 127 2.071 1.186 127.5 1.234 1.228 127.5 1.215 1.204 127.5 1.493 1.155 127 1.385 1.154 127.5 1.565 1.138 127.5 1.205 1.309 127.5 2.525 1.482 127.5 3.489 1.657 127.5 1.035 1.405 127.5 1.478 1.285 127.5 3.605 1.118 127.5 1.36 1.23 127.5 1.169 1.303 127.5 1.643 1.35 127.5 0.975 1.449 127.5 1.076 1.352 127.5 1.459 1.354
Event 6 Stretching Start End factor (Ma) (Ma) â ä
Tunja 133.5 Tunja-1 132 Bolivar-1 Corrales-1 132 98 Caqueza 144 98 " " " * 144 98 Nazareth 131.5 98 Paz de Rio 144 98 Servita 132 98 Aguazul 142 98 Chita 139 98 Cocuy 139 98 Guateque 138 98 Labateca 132 98 Medina-1 144 98 Mojicones-1 139 98 Pajarito 142 98 R Cusay 136.5 98 San Luis de Gaceno 144 98 " " " * 144 98 Sogamoso 136.5 98 Apiay-4P 85.9 Arauca-1 98 Ariari-1 98 Arimena-1 98 Balastrera-1 98 Cachama-1 98 Camungo-1 98 Caño Barulia-1 98 Caño Bravo-1 98 Caño Cumare-1 98 Caño Duya-1 98 Caño Garza-1 98 Caño Limon-1 98 Caño Rondon-1 98 Casimena-1 98 Centauro-1 98 Centauro Norte-1 98 Chaparrito-1 98 Chichimene-1 98 Corocora-1 98 Cravo Este-1 98 Cusiana-1X-2 112.3 112.2 1.095 1.025 98 El Palmar-1 98 Entrerrios-1 98 Garibay-1 98 Guarimena-1 98 Guarrojo-1 98 Joropo-1 98 Kioskos-1 98 La Cabaña-1 132 127.5 1.478 1.285 98 93 1.581 1.161 98 La Flora-1 98 La Maria-1 98 La Punta-1 98 La Tortuga-1 98
# Numbers indicate Cretaceous sub-basins as shown i n Fig. 2.18. * Modelled using Triassic and Jurassic actual or inferred stratigraphy. Table 2.4 b: Continued
41
93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 85.8 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93 93
1.205 1.29 1.002 1.022 1.07 1.166 1.096 1.037 1 1.092 1.581 1.103 1.157 1.108 1.233 1.02 1.277 1.06 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.581 1 1 1 1
1.054 1.129 1.032 1.043 1.07 1.072 1.098 1.037 1 1.052 1.161 1.103 1.109 1.104 1.076 1.055 1.125 1.06 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.161 1 1 1 1
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Basin compart sub-basin # 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
Stratigraphic column
Event 4 Stretching Start End factor (Ma) (Ma) â ä
Event 5 Stretching Start End factor (Ma) (Ma) â ä
Leticia-1 Los Teques-1 Los Trompillos-1 Manacacias-1 Manacacias-2 Maremare-1 Metica-1 Morichito-1 Negritos-1 Pajaropinto-1 Piriri-1 Planas-1 Pomarroso-1 Remache Sur-1 Rubiales-1 San Joaquin-1 San Juan-1 Sardinas-2 ST-0-04 ST-CN-8 Stella-1 ST-GU-15 ST-GU-19 Surimena-1 Surimena Norte-1 SV-9 Tierrablanca-1 Tocaria-1 Upia-1 Valdivia-1 Voragine-1 Yucao-1
Event 6 Stretching Start End factor (Ma) (Ma) â ä 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1 98 93 1 1
# Numbers indicate Cretaceous sub-basins as shown in Fig. 2.18. * Modelled using Triassic and Jurassic actual or infer red stratigraphy. Table 2.4b; Continued
4.2. TECTONIC SUBSIDENCE DURING CRETACEOUS TIME The location of stratigraphic columns with Cretaceous section is shown o n Figure 2.2. Figure 2.21 il lustrates the great thickness of the Cretaceous section in some parts of the basin. Figure 2.22 shows the tectonic subsidence curves during Cretaceous time. Curves have been groupe d according to their similar shape and relative location within the basin into several basin compartments as indicated in Figure 2.18. 4.2.1. Basin compar tments
Tectonic subsidence curves ( Fig. 2.22) and restored thickness maps (Figs. 2.13, 2.15 an d 2.17) indicate that several basin compartments existed ( Fig. 2.18). Distin ct compartments had different subsidence histories, and being bounded by f ormer extensional fault systems, as suggested by local geological evidence (Figs. 2.23 and 2.24), and constrained further by lateral thickness changes at opposite sides of faults (Figs. 2.13, 2.15 and 2.17). Most of these faults have reversed their relative movement during the Cenozoic (Fa bre, 1987; Colletta et al.,1 990; Cooper et al.,1 995; Figs. 2.23 a nd 2.24). In the northern part of the EC comparison of Cretaceous subsidence in a W-E transect A (Figs. 2.13 a nd 2.18) r eveals the following trends:
42
0
Chapter 2 0
1
r
S
1 . .
.
. .
8 0
S
C
A
1
1 W 9 0
1 2
) c e s ( T W T
43
3
4
e d n t a h c f i o h s p s a e r n g i k t c a i r h t s t t e a c e a r f g r u e s h e t l e b a t o l i N a v . e A n . i l n i s c a i b m s i b u s s e a e l c r o a h m w a e n i h t d n u g n C l o e a h t t u f o o t s r p a p o r c n r . e p t s e G a w t e e l h i t l n V i s n u s i o o i e t n c c o a e i t s e t c r c i s C e m s e s i i h h e t s T f . o a d f e n o t o a i g t c i n d a i n c o w i a L r s . 2 d i n . l e o 2 o n t t r i e i r L n c e s u : o g c s i 1 2 u l F . o 2 a r n e i u e c t r a c t w u u g t r o r e i F s C h s
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
(1)
Two sub-basins: the Cocuy (region 9) and Tablazo (region 3) rift sub-basins separated by the less subsiding Santander-Floresta block (region 7). (2) A regional westward decrease in tectonic subsidence with a maximum in the Cocuy sub-basin (region 9) and a minimum in the Middle MV (region 2), suggesting a regional half-rift geometry for the whole basin. A comparison of the Cretaceous subsidence along a W-E transect B ( Figs. 2.13 and 2.18) in the southern part of the Cordillera at the latitude of Bogota, reveals the foll owing: (1) The existence of a single rifted basin, the Cundinamarca sub-basin (region 4, Fig. 2.18). Thickness variations of the drilled section in the axial region with respect to the sections exposed along the flanks of the Cordillera within specific time intervals, suggests an absence of less subsiding ridges or highs in the axial region. Gravity models calculated by Calvache and Muñoz (1984) and Kellogg and Duque (1994) support the latter statement. (2) Tectonic subsidence during the earliest Cretaceous (Berr iasian to Hauterivian) was maximal in the eastern side of the Cundinamarca sub-basin (region 8, Fig. 2.18), indicating that a first stretching event mainly affected the eastern Guaicáramo normal f ault system. Later during the Aptian, however, subsidence was maximal in the western side of the Cundinamarca sub-basin (region 4, Fig. 2.18) suggesting that a second stretching event mainly affected the western Bituima fault system. The total tectonic subsid ence during the whole Cretaceous was slightly greater in the western side of the basin (region 4). A thick Cr etaceous section in the western side of the Cundinamarca sub-basin is illustrated i n Figure 2.21. In the easternmost LLA area (region 10, Fig. 2.18), sedimentation started during Late Cretaceous and the total tectonic subsidence during the Cret aceous was small (100 to 200 m) compared to the EC and MV. Subsidence probably was produced by flexural thermal subsidence (Watts et al ., 1982), and by water loading due to increase in palaeo-water depth. Total tectonic subsidence in the Upper MV (region 6, Fig. 2.18), where marine sedimentation started in Aptian time (Fig. 2.16), is significantly smaller than that of the EC and Middle MV (region 2). 4.2.2. Cretaceous fast subsidence events
Basin com partmentalisation indicates differential tectonic subsidence of fault-bounded blocks( Figs. 2.23 and 2.24). Using the single instantaneous stretching model of McKenzie (1978), Hébrar d (1985) and Fabre (1987) proposed that a lithosphere-stretching event during early Cretaceous was followed by thermal subsidence. I identified several rapid subsidence events ( Fig. 2.22). In the following section I will attempt to test if these different rapid tectonic subsidence e vents were structurally controlled by documented faults and correlate with evidence of tectonic extensional movements, to infer if they were produced by lithosphere extension, or by other process. Table 2.4b hi ghlights the different rapid subsidence events as interpreted from the subsidence curves for each stratigraphic column. Most subsidence curves allow identification of several steps of increased subsidence. Comparison between different subsidence curves indicates that each rapid subsidence event was not synchronous within the whole basin. The following events of rapid subsidence have been identified for the Cretaceous ( Fig. 2.22): Berriasian to Hauterivian event (altho ugh variable in different stratigraphic columns comprised between 144 to 127 Ma). It occurr ed in the area of the EC and i s best repr esented in its eastern flank. Subsidence curves ( Fig. 2.22) and thickness maps (Figs. 2.13 and 2.15) show evidence for a wide (>180 km wide) asymmetrical half-r ift basin divided by the Santande r-Floresta high. Maximum tectonic subsidence was assoc iated with the pre-Guaicáramo normal master fault system that was the eastern boundary of the rift ( Figs. 2.31 and 2.32). A second order half graben was located at the current location of the wester n flank of the EC. This minor half graben probably was associated with a palaeo-normal fault system a pproximately following La Salina-Bituima fault system that was its western border (Figs. 2.13, 2.15, 2.31 and 2.32). To the south there was only one depocenter , limited in the south by a NW-SE vertical transfer fault (Fabre, 1987; Sarmiento, 1989; Figs. 2.31 and 2.32). If plate-tectonic interpretations by Pindell and Erikson (1993) are accepted this fault probably was a rolongation of a transform fault of the proto-Caribbean ridge system (Geotec, 1992).
44
Chapter 2
Geotec (1992) suggested that NW-SE trending faults formed a gr aben with this o rientation in the northern Central Cordillera. However, subsidence in this area is small ( Figs. 2.18 a nd 2.22).Early Cretaceous turbidites at both flanks (Murca Fm, Sarmiento, 1989; Moreno 1990b, 199 1; Cáqueza group, Pimpirev et al ., 1992; Branquet, 1999) of the extensional basi n (Fig. 2.12) can in this situation be taken as evidence of tectonic instability associated with normal fa ult movement. Branquet (1999) presented outcro p and seismic evidence of Cretaceous normal faulting. Normal faults imaged on seismic sections (Figs. 2.23 a nd 2.24) confirm extensional tectonic movements that attest this rapid subsidence event was produced by lithosphere stretching. Contrary to the Triassic and Jurassic rifting, magmatic activity was reduced during the Early Cretaceous. Evidence for Early Cretaceous magmatism is limited to (1) small mafic igneous intrusions described by Fabre and Delaloye (1983) and Moreno and Concha (1993; Fig. 2.31) and (2) some volcanic input within Cretaceous shales (Rubiano 1989; Villamil, 1994). Aptian to Early Albian event (comprised between 121 – 102.6 Ma). This fast subsidence event occurred at the current location of the EC and MV. However, it is best represented in the southern part of the western flank of the EC and the Upper MV, indicating asymmetry in the basin. During Bar remian to Aptian time the basin was extended toward the south in the Upper M V (Figs. 2.16, 2.17, 2.22). Turbiditic deposits of Aptian age (Socota Member described by Polan ía and Rodríguez, 1978; S armiento 1989) can in this case also be tak en as evidence of tectonic instability associated with this rapid s ubsidence event. The isopach map ( Fig. 2.17) suggests that a master normal fault system, located appro ximately at the present day B ituima Magdalena fault system, was active ( Figs. 2.33 an d 2.34). In the area of the Upper MV a normal palaeo-Chusma fault system was probably also active (F igs 2.33 and 2.34). This fast subsidence event is interpreted as produced by lithosphere stretching. Evidence o f magmatic activity within the basin is limited to some small mafic intrusions (Fabre and Delaloye, 1983), and minor volcanic input (Rubiano, 1989; Villamil and Arango, 1998). Some zircon fission track ages (see Chapter 3) from the Mérida Andes (Kohn et al ., 1984), Sierra de Perijá (Shagam et al ., 1984), and the Garzón Massif (Van der Wiel, 1991) seem to correlate with this event. Van der Wiel (1991) interpretated this ages as related to an orogenic event that affected the whole northwestern corner of South America between 100 and 80 Ma. However this is in contradiction with strong stratigraphic evidence of a subsiding basin at these localities. Instead if this correlation is valid probably the zircon ages reflect local uplift of faulted blocks located at rift margins, as demonstrated using fission track data from a number of rift basins by Van der Beek (1995). This author explains rift margin uplift by mechanical support of rift flanks resulting from an upward state of flexure. Cenomanian event (98 to 93 Ma) occurred in the easter n flank of the EC. No major geological evidence has been reported for a tectonic event at this time ( Fig. 2.22) in the EC. Based on thickness changes across the Cusiana-Támara fault system (Fig. 2.34), Cooper et al . (1995) suggested that normal fault movement took place during the Ca mpanian (73 to 80 Ma) and possibly earlier time. During this time some small mafic intrusions were emplaced in the most subsiding parts of the basin (Fabre and Delaloye, 1983). Van der Wiel (1991) reported in the Garzón Massif zircon fission track ages comprised between 113 to 85 Ma (including Cenomanian) and interpreted them as an orogenic event. However, in the area of extensional basins the zircon ages probably reflect cooling and uplift of rift shoulders (after lithosphere stretching, c.f.V an der Beek, 1995) or intrusion of small mafic bodies. Some Late Cretaceous fission track ages from the Central Cordillera (Gomez et al., 1999; Toro et al., 1999) indicate initial uplift of this mountain range giving support to the hypothesis of a Late Cretaceous magmatic arc in the Central Cordillera. During the late Cenomanian-Turonian, the global sea-level maximum (Haq et al ., 1987; Villamil and Arango, 1998) correlates with this subsidence event, suggesting that increase in subsidence was driven by water load during maximum sea-level. However, it is difficult to rule out the possibility of some tectonic lithosphere extension component, especially for normal fault activity in the Cusiana-Támara Fault between the EC and the LLA basin (Cooper, et al., 1995).
45
e
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes c
1 . C E N TR A L C O R D I L L E R A Age ( M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60
50 40 0
San Pabl o Abe jor ra l 50 0
San Felix
T 2 . MI D D L E M A G D A L E N A V A LL E Y Age ( M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
2 S an
15 0
14 0
12 0
13 0
11 0
10 0
90
80
70
60
50
40 0
Luc as
e 50 0
2 C a s c a j a 2 E l l e s S u d a
1000
n M o r r o c o ya l
2 Cimitarra 1500
Sabalo-1k 2000 Casabe- 199 Infantas- 1613 Llanito-1 Lebrij a
T
Cascajales-1 2500 Tabl azo 3000
e
3 . T A BL A Z O S U B - B A S I N Age ( M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60
40
50
0
3 Chima
3 Onzaga
3 M e d i o s
3 La
Los Santos Vadoreal
R u s i a
50 0
Simac ota Chima
1000
T a b l C i m i t a r r a a z o
M edios
Ar ca buco 1500
T Villa de Leiva
2000
4 . C U N D I N A M A R C A S U B - B A S IN Age ( M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60
50
40 0
Q u i i l p e 50 0 Chitasuga- 1 Suesc a N-1 1000 Suba-2 Suesca - 1 Sutamarchan L a Simijaca Tabio C 1500 a l e Villeta r a
Yacopi 2000
Figure 2.22: Tectonic subsidence curves from the whole Mesozoic sedimentary record. Horizontal axis represents age in Ma. Vertical axis represents tectonic subsidence in meters obtained from backstripping a nalysis. Ver tical shaded strips represent fast subsidence events. Numbers refer to basin compartments shown in Fig. 2.18. Note also the vertical bars representing the fast tectonic subsidence events.
46
T
Chapter 2 c (c
5. SOU THERN CUND INA MARC A SUB-BASIN Age (Ma ) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
e 14 0
13 0
12 0
11 0
10 0
90
80
70
60
50
40 0
)
50 0 Fusagasuga Apu lo
T
1000
6. UPPER MA GDA LEN A VALLEY Age (Ma ) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0 16 0 15 0 14 0 13 0 Alp uja rr a Prado Coello Luisa Neiva Ata co
12 0
11 0
10 0
90
80
70
60
50
40
0
Itaibe Girardot
Guataqui Prado Coello Neiva Alp uja rr a Ortega Q Cal ambe Q El C obre
Payande Q Olini R S aldaña Q Olini
e
50 0
1000
c
1500
T
0
)
7. SANTAND ER FLO RES TA MASSIF Age (Ma ) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60
50
40
Floresta Guaca
Matanza Tibasosa Tunja Corm ic hoque Tunja-1 Chivata
50 0
ie 1000
T
8. EASTERN CUND INAMARCA A ND WESTERN C OCO UY SU B-BASINS Age (Ma ) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60
50
40 0
8 Buenavista
50 0
Cormichoque-1
Paz de Ri o
Nazareth Servita Bolivar- 1000 Corrales Caqueza
c
T
1500
9. COC UY SUB-BA SIN Age (Ma ) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
9 Bata
13 0 12 0 11 0 10 0 L a b a t e c a M o j i c o n e s
S o g a m o S a n L
R C us ay
s o
C h i t a
90
80
70
60
50
40
0
50 0
P a j a r i t o
1000 Agua zul
c
u i s d e G a c e n o
1500 Cocuy
T
Guateque 2000
Figure 2.22: Continued
47
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes 10. LLANOS OR IEN TALES Age ( M a) 25 0
24 0
23 0
22 0
21 0
20 0
19 0
18 0
17 0
16 0
15 0
14 0
13 0
12 0
11 0
10 0
90
80
70
60 50 40 0 La Tortuga-1
Cañ o Limon-1
Arauc a-1
50 0
Figure 2.22: Continued Maastrichtian to Paleocene event (although
variable in different columns, it is comprised between 68 and 54.8 Ma). This fast subsidence event affected the axial part of the EC, its eastern flank and locally the westernmost part of the LLA area. This event correlates in time with deformation and uplift in the Central Cordillera (Jaramillo, 1978, 1981; Cooper et al ., 1995). Some authors (e.g. Kluth et al., 1997) recognized or suggested normal faulting in the Llanos area approximately at this time.
5. CORRELATION OF FAST SUBSIDENCE EVENTS WITH MAGMATIC, EUSTATIC AND PLATE-TECTONIC EVENTS Below I explore the timing and nature of changes in subsidence, in the context of changes in plate-tectonic regime and eustasy.
5.1. CORRELATION BETWEEN FAST SUBSIDENCE EVENTS AND SUBDUCTIONRELATED MAGMATIC ARCS The inferred Mesozoic stretching events seem to correlate in time with reduced magmatic activity in the Central Cordillera ( Fig. 2.25 m odified from Aspden et al ., 1987 and Guillande, 1988). If the calc-alkaline (Alvarez, 1983) plutonic belts of the Central Cordillera were developed as subduction-related magmatic arcs during Mesozoic times, as suggested by Aspden et al ., (1987), the extensional basins behind them may be interpreted as backarc basins. Extensional backarc basins are developed when the velocity roll-back, due to fast subduction, exceeds the oceanward convergence velocity of the overriding plate (Dewey, 1980; Cross and Pilger, 1982; Royden, 1993a,b). If magmatic arc activity decreases with the oceanward convergence velocity of the overriding plate, during times of reduced magmatic arc activity a constant roll-back velocity would exceed the oceanward convergence velocity of the overriding plate, increasing extension and subsidence in the backarc region. According to Aspden et al. (1987) the Triassic magmatic belt was controlled along strike-slip faults. Evidence of the latter is presented by Restrepo-Pace (1995). Jurassic calc-alkaline plutonism along the Central Cordillera (Restrepo et al., 1991) was interpreted by Aspden et al ., (1987) as a subduction-related magmatic arc. However, Cretaceous plutonism is sporadically developed only in the northern part of the Central Cordillera (Restrepo et al., 1991) whereas it is very extensive in Peru (Cobbing, 1982 in Aspden et al ., 1987). Aspden et al . (1987) suggested oblique convergence and an offset in the subduction zone along a major NE-SW transform fault to account for the notable absence of Cretaceous plutonism in southern Colombia and Ecuador. Triassic-Jurassic magmatism in the Santander Massif (Ward et al., 1973; Restrepo-Pace, 1995) suggests a short magmatic arc segment (Restrepo-Pace, personal communication) that correlates with the small subsidence in this region. Arcs tend to be associated with uplift due to crustal thickening and thermal and physical effects of rising magma, or with reduced subsidence (Smith and Landis, 1995).
48
T
Chapter 2 NW
Seismic Profile ME94-1460
SE t l u a f r a l a c a u g A
Carbonera Fm.
0
1. 0
Carbonera Fm. . F m o r d a M i r
2. 0
Mira dor
T W T
Fm .
K
Mira dor
Fm .
3. 0
Basement
Basement
4. 0
Basement
a.
5. 0 0
NW
t l u a F a i l a s e T
5 km
SE
Seismic Profile ME92-1260
0 t l u L e a F o n F m a r . a l c C a a u r b g o n A e r
. F m n o L e r a n e b o r a C
. F m
a
F m 1. 0 .
T W T F m . d o r M i r a
. F m o r d a M i r
K
2. 0
K
K
3. 0
n t a m e B a s
4. 0
b. 0
5 Km
Figure 2.23: Seismic sections in the Medina foothills area, along the eastern border of the Cundinamarca sub basin. Note normal fault evidence during Cretaceous (K) in the Guaicáramo palaeo-fault system along the eastern border of the basin and contractional inversion of Cretaceous extensional faults occurr ed during Palaeogene time as evidenced by lateral changes of thickness of the Palaeogene Carbonera Fm. (see Chapter 3). Note also the thickness changes in the Cretaceous sedimentary fill (from Linares, 1996). Location of t his seismic line is shown in Figure 2.2.
49
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
Seismic line DM87-1470e
SE 5 km
Arru gas -1
SE Peñ a d e O ro-1 San Lui s-1 2
0.0
0.5
1.0
1.5
A.
K
2.0
) c e s ( T W T
N
0.5
2 km
Seismic line CL861393 900
800
700
600
500
400
300
200
SE
100
1.0
) 1.5 c e s ( T W T
K
2.0
2.5
B.
3.0
Figure 2.24: Seismic sections in the Middle Magdalena Valley. Note normal faulting evidence during Cretaceous (K) in the western border of the basin A. Seismic line DM87-1470e. Bottom of Cretaceous is the light green lowermost reflector affected by normal faults. Note the in creasing thickness of Cretaceous (K) toward the SE and normal faults (modified from Olaya and Serrano, 1998). B.S eismic line CL861393. Bottom of Cretaceous is the purple lowermost reflector affected by normal faults. Note the lower-middle? Eocene unconformity truncating Cretaceous strata. (From Mantilla, 2000). Location of these seismic lines shown in figure 2.2.
50
Chapter 2 Panama dates 74 Oca Fault
78
5 0
Caribbean Sea
15
s
5
Venezuela
P a n a m a
l t u a F tu O
8
g
Pacif ic Ocean
t l u a F
c a u C a
0
Ecuador
t
B u S c a a r a n m t a a n M g a a r F t a a u tl
u l t Fa n a ti l e s a t l P u
a F t o l m u a r a l a F a c r i i a m e u a t a G P R o a
4
Zone V
10
15 Zone IV
10 5
t
0 15 5 10
3
2
1
Zon e III
5
Zone V Zone IV
4
N
0
Zon e III
10
Zo ne II
5
Zone I
Zo ne II
0
Triass ic plutons
5
Major f ault zones/sutures
0
Zone I 0
50
100 150 200 Pluton ages (Ma)
250
0 3 50
L i t h o s p h e re stretching e v e n t s.
Periods of intense m a g m a t i c a c ti v it y
10 0 5 4
A
15 0
P e r i o d s o f i n t e rm e d i a t e m a g m a t i c a c ti v it y
3 2
20 0
1
25 0
b
P e r i o d s o f sl i g h t o r n o m a g m a t i c a c t i v it y
0
50 10 0 C u m u l at i v e n um b er o f r a di o m e t r i c a ge determinations
Figure 2.25: Event correlation between lithosphere stretching in the area of the Eastern Cordillera and magmatic activity in the Central Cordillera. 1 Triassic event. 2 Late Triassic-Early Jurassic event. 3 Middle Jurassic event. 4 Early Cretaceous Berriasian-Hauterivian event. 5 Aptian event. This correlation should be considered preliminary because original data are heterogeneous. 94% of data are K-Ar (biotite, hornblende, muscovite or whole rock) and 6% of data are Rb-Sr (hornblende/biotite or whole rock). A. Left panel: Principal structural/plutonic zones of Western Colombia. Right panel: Age distribution of Mesozoic and Cenozoic plutonic activity in Western Colombia (modified after Aspden, 1987). B Cumulative histogram of radiometric ages of plutonic bodies in Colombia (modified after Guillande, 1988). Periods of intense magmatic activity are characterized by a rapid increase in the cumulative number of radiometric age determinations for a time interval (low slope of the curve).
51
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
5.2. CORRELATION BETWEEN FAST SUBSIDENCE EVENTS, PLATE-TECTONIC EVENTS AND EUSTATIC EVENTS Examination of Caribbean plate-tectonic interpretations (Pindell and Dewey, 1982; Burke et al., 1984; Duncan and Hardgraves, 1984; Ross and Scotese, 1988; Jaillard et al., 1990; Pi ndell and Barret, 1990; Pindell and Erikson, 1993; Meschede and Frisch, 1998 ; Figs. 2.2 and 2.3) and sea-level curves (Haq et al ., 1988; Villamil and Arango, 1998) suggests the M esozoic ra pid subsiden ce events correlate with: Triassic event. Reduced magmatic activity in the Central Cordillera (Fig.2.25) was then related to oblique subduction (Aspden et al., 1987). The alternative hypothesis of intracontinental rifting related to separation between South and North America, is applicable to Venezuela and probably nor thern Colombia. Late st Triassic-Middle Jurassic event. Reduced magmatic activity in the Central Cordillera (Fig. 2.25) may be interpreted as backarc extensio n (Fig. 2.3). According to Aspden et al . (1987) this reflects a change in convergence of the subducting p late from NNW/SSE to NW/SE. Middle Jurassic event. Reduced magmatic activity in the Ce ntral Cor dillera (Fig. 2.25) may be correlated with backarc extension related to high-angle subductio n (Fig. 2.3, Aspden et al., 19 87). According to Pindell and Erikson (1993) and Meschede and Frisch (1 998) interpretations, opening of the proto-Caribbean started in Northern Colombia. Berriasian-Hauterivian event. According to the Cretaceous passive margin interpretation (Pindell and Erikson, 1993), active opening of the proto-Caribbean was occurring north of Colombia and also west of the palaeo-Central Cordiller a (Fig. 2.4). If such an interpretation is valid for the Berriasian to Hauterivian, stretching in the stu dy area produced a failed-rifted arm related to a major opening of the proto-Caribbean oceanic basin. The alternati ve plate-tectonic reconstruction attributes Early Cretaceous rifting of the palaeo-EC to a backar c basin contemporaneous with reduced magmatic activity in the Central Cordillera (Fig. 2.4). R educed magmatic activity in the Central Cordillera may be also associated with accretion of the Amáime Terrane along the Romeral Fault west of the Central Cordillera (Fig. 2.5) at about 125-130 Ma (Feininger, 1985, 1986; Aspden and McCourt, 1986). Widespread dynamother mic metamorphism occurs in the Central Cordillera with emplacement of blueschists (Aspden et al., 1987). The following evidence supports the hypothesis of a back-arc basin located behind a partially emerged, less subsident palaeo-Central Cordillera (magmatic arc?): (1) The presence, in the western part of the Cundinamarca sub-basin, of Lower Cretaceous sandstones with abundant volcanic lithic fragments and feldspar derived from a western detrital source area as indicated by palaeocurrent data (Murca Fm. and Útica Sandstone; Sarmiento, 1989; Moreno, 1990b, 1991). (2) The presence of progressive westerly onlap terminations of the Cretaceous carbonates on the basement, observed in seismic lines, in the wes tern border of the Cesar Valley, in northern Colombia (in Mesozoic times part of the EC basinal ar ea, Fig. 1.1; Audemard, 1991). (3) Stratigraphical and petrographical eviden ce suggesting that during Berriasian (?) to Valanginian time clastic sediments near San Felix in the western flank of the Central Cordillera (between the Romeral and the Palestina Fault s, Fig. 2.5) came from erosion of uplifted areas with metamorphic rocks and small tectonic blocks with plutonic rocks (Rodríguez and Rojas, 1985). Rodríguez and Rojas (1985) identified west-verging thrust faults post mid-Albian/pre-Miocene (but according to these authors fault activity probably started during Albian time). (4) The presence of some Cretaceous volaniclastic rocks in the Central Cordillera (Rodríguez and Rojas, 1985), made up of mixture of pyroclastic and epiclastic fragments probably derived from a magmatic arc. (5) Relatively high concentration of volcanogenic clay minerals in Hauterivian-Barremian (0 to 30%), middle Albian (0 and 21%), and Turonian (6 and 9 %) shales of the Villeta Gp (Rubiano, 1989) and Valanginian-Hauterivian Rosablanca Fm (Moreno, 1989 in Rubiano, 1989; Moreno, 1990a) in the Cundinamarca sub-basin. Thin beds of volcanogenic clays or bentonites within the Cenomanian- Turonian stratigraphic interval (Villamil and Arango, 1998) and the Salada Member of the La Luna Fm (Patterson, 1970 in Rubiano, 1989), as well as subaqueous volcanic volcanic tuffs within La Frontera and La Luna Fms in the MV (Restrepo-Pace, personal communication).
52
Chapter 2
(6) Jurassic (185 Ma) and Cretaceous ( ≤ 77 Ma) zircon fission track ages from the Central Cordillera (Toro et al., 1999; Gómez et al ., 1999) evidencing uplift. In the Ecuadorian Andes (Rivadeneira, 1996) and the Central Cordillera (Rodríguez and Rojas, 1985) these authors have suggested uplift or deformation during Late Cretaceous time. Late Cretaceous (85 to 113 Ma) zircon ages in Venezuela and the Garzón Massif suggest a tectonic-thermal event interpreted by Van der Wiel (1991) as an orogenic event. However in the area of the Mesozoic extensional basins it is difficult to assume orogenic uplift. Probably these data indicate local uplift of faulted blocks located at rift margins ( c.f.V an der Beek, 1995). However this hypothesis needs further confirmation. (7) Cretaceous igneous intrusions, as the Late Cretaceous Antioquian Batholith and others of Early Cretaceous age in the Central Cordillera ( e.g. San Diego, Cambumbia and Mariquit stocks; Restrepo et al., 1991) which define a magmatic arc. However such a magmatic arc is not well defined. Aptian-Early Albian event. Pindell
and Erikson (1993) hypothesized that during Aptian the western border of Colombia was a passive margin. Spreading west of Colombia ceased and the proto-Caribbean lithosphere began to subduct westwar d under the Amáime-Antilles arc that was approaching the western margin of northern South Ame rica (Fig. 2.4). If this interpretation is valid, stress changes due to the above-mentioned plate-tectonic s changes could have triggered stretching in the study area. The alternative plate-tectonic hypothesis of Meschede and Frisch (1998) also assumes possible extinction of spreading of the proto-Caribbean lithosphere northwest of Colombia and the beginning of subduct ion of the Farallon/Pacific plate under the Panama-Costa Rica arc west of Colombia ( Fig. 2.4). Such plate-tectonic reorganisations could have induced lithosphere stretching in the palaeo-EC basin. Bourgois et al . (1982a,b; 1987) proposed that obduction of oceanic terranes took place over the Central Cordillera. Cenomanian event. This age is included within the 100 to 80 Ma time interval proposed by Duncan and Hardgraves (1984) and Hill (1993) for the formation of the Caribbean plate by partial melting within the initial plume head of the Galapagos hotspot. The plate-tectonic interpretation of Pindell and Erikson (1993) proposes that during the Cenomanian Colombia remained as part of a passive margin without major changes in its plate-tectonic configuration. An alternative platetectonic interpretation (Nivia, 1987; Meschede and Frisch, 1998) proposes that the formation of the basalt plateau in the Caribbean area, west and northwest of Colombia, thickened the proto-Caribbean crust. If such an interpretation is valid, buoyancy of the thickened Caribbean crust could have impeded its continuous subduction under the Central Cordillera located at the continental margin. Consequently, backarc subsidence is not favoured in this scenario. Late Cenomanian-Turonian global sea-level maximum correlates with this subsidence event, suggesting that increase in subsidence was driven by water load. However, the event affected only the eastern flank of the Cordillera, where the maximum thickness of Cretaceous is present. While the maximum flooding surface for the Cretaceous sediments of the EC is the Cenomanian-Turonian boundary (Villa mil and Arango, 1998), it is Campanian in the eastern LLA (Fajardo et al., 1993; Cooper et al., 199 5; Figs 2.7 to 2.11). If subduction of the Caribbean thick and buoyant lithosphere under South Am erica was inhibited, it probably exerted horizontal stresses on the northwestern margin of South America. Horizontal stresses can induce local flexural lithosphere bending, which is maximal where the lithosphere is weakest (Cloetingh, 1988; Cloetingh and Kooi, 1992). This process probably enhanced the relative sea-level rise, creating a maximum Cenomanian-Turonian marine flooding surface in the depocenter of the EC, characterized by weak lithosphere due to earlier stretching. In contrast, at that time horizontal stress produced a submarine shallow water depth bulge in the LLA, which partially compensated the maximum eustatic signal. Maastrichtian-Paleocene event. All plate-tectonic interpretations agree that during latest Cretaceous and probably Paleocene, the accretion of the Western Cordillera oceanic terranes along the Cauca-Patia Fault occurred, producing deformation and uplift of the Central Cordillera. According to Cooper et al . (1995) loading of the Western and Central Cordilleras led to the development of an early pre-Andean flexural foreland basin in the area of the EC. However, subtle uplift and erosion in the Tunja-Sogamoso axial region and initial inversion of the extensional basin have also been proposed by Fajardo-Peña (1998). Increased subsidence in the axis of the Cundinamarca sub-basin (Sabana de Bogotá) could be the result of increased horizontal
53
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
compressional stress (Cloetingh, 1988; Cloetingh and Kooi, 1992) associated with collision of the oceanic terranes of western Colombia and deformation and uplift of the Central Cordillera. Development of normal faults in the Llanos area ( e.g. Kluth et al. 1997) could be the result of local tensional stresses in the flexural bulge developed.
6. FORWARD MODELLING OF BASIN EVOLUTION Subsidence analysis has allowed the identification of seven events of rapid tectonic subsidence followed by slower subsidence rates. The first five events of rapid tectonic subsidence, which can be clearly correlated with tectonic activity, are interpreted as stretching events followed by periods of slower thermal subsidence. In order to quantify horizontal extensional movements responsible for the observed subsidence and establish a quantitative framework for the pulsating rift evolution of the lithosphere during Mesozoic basin formation, Sarmiento and Van Wees ( in prep.) have quantified extension rates by forward modelling of tectonic subsidence. They used an “automated” forward modelling technique (Van Wees et al., 1996b), which will be briefly explained below.
6.1. NUMERICAL MODEL The forward modelling approach is based on lithospheric stretching assumptions (McKenzie, 1978; Royden and Keen, 1980). The extension factor δ is used for crustal stretching and β for subcrustal stretching. For the thermal calculations, a 1D numerical finite-difference model was used, which allows incorporation of finite and multiple stretching phases. In order to handle a large number of wells and stretching phases in the forward model, a numerical technique is applied (Van Wees et al., 1996b), which automatically finds best fit stretching parameters for (part of) the subsidence data. In this procedure the timing and duration of the rift phase must be specified, whereas best fit stretching values are found by search ing for the minimum of the mean square root F of the deviation in predicted and observed subsidence ( Fig. 2.26), as a function of δ, β:
F (δ , β ) =
1 num
i = num
2 ( ) − s s ∑ , , p i
oi
(2.1)
i =1
where num is the number of subsidence data used in the fitting procedure and s p,i, so,i are predicted and observed subsidence values respectively. For a rift phase, either uniform lithospheric stretching (δ = β) (McKenzie, 1978) or two-layered stretching can be used ( δ ≠ β) (e.g. Royden and Keen, 1980). For uniform stretching, the solution of Eq. 1 requires that at least one observed subsidence data point is given after the onset of rifting, whereas the two layered stretching requires at least two data points. For polyphase stretching, the fit is accomplished in sequential order. Initially, using an initial steady state thermal and compositional lithospheric configuration (cf. McKenzie, 1978), stretching parameters of the first phase are determined by fitting data points in the syn-rift and post- rift time interval up to the onset of the following phase.
54
Chapter 2 A g e (M a ) 1
20 0
2 1
t
2
10 0
0 0 20 0 40 0 60 0
sp
p
80 0
1
s o1
so2 s p2
p
1000
sp3
1200
s o3
1400
s
1600 1800 2000
su b s i d e n c e d a t a f i t t e d b y p h a se 2
su b s i d e n c e d a t a f i t t e d b y p h a se 1
Figure 2.26: Outline of the forward modelling technique. Explanation in the text (from Van Wees et al., 1996b).
Subsequently, using the perturbed lithosphere configuration predicted at the onset of the second rift phase, stretching parameters of this rift phase are determined using subsidence data from its syn-rift and post-rift time intervals up to the next rifting phase. In the solution procedure it is assumed that values for δ and β can both be found in an interval from 0.2 to 10. This covers the whole spectrum of realistic parameters for crustal and subcrustal thickening and thinning. In case of high tectonic uplift or subsidence in this post-rift interval, it may be difficult to obtain a good fit for the subsidence, since subsidence and uplift rates are determined by the thermal relaxation of the subcrustal lithosphere. Maximum uplift and subsidence rates are for β = 0.2 and β = 10. Such predictions, implying extreme thermal perturbations of the subcrustal lithosphere, should be interpreted with great care. u
6.2. MODELLING PROCEDURE In the fitting procedure initial lithospheric configuration and thermal parameters are adopted as listed in T able 2.2. T o fit the data it was assumed that each observed phase of rapid tectonic subsidence should corres pond with a stretching phase in the forward model. For these phases I adopted a two-layered stretching model of the lithosphere ( δ ≠ β) in order to obtain the highest degree of freedom in modelling subsidence data. However, I prefer to use a uniform stretching model (δ = β) for those cases where uncertainty exists in estimating stretching factors due to a small number of data points, or where there are relatively large age uncertainties, as is the case of the Triassic and Jurassic sedimentary record. Using the starting and finishing times previously determined for the stretching events, I calculated the lithosphere stretching factors that would produce theoretical subsidence curves
55
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
similar to those observed. For the forward modelling, I included for most modelled locations the complete Mesozoic sedimentary section since the Triassic, even in those columns where the preMesozoic section is probably deep and does not crop out. In these cases I used thicknesses interpolated from the isopach maps. In other cases, I only studied the Cretaceous subsidence. As the lithosphere affected by Triassic-Jurassic rifting would behave differently to those unextended areas, I applied a reduced crustal and lithosphere thickness in the previously rifted area for the forward modelling of the Cretaceous subsidence. The reduced crustal and lithosphere thicknesses were calculated using lithosphere stretching factors calculated from Triassic-Jurassic data.
6.3. MODEL RESULTS: STRETCHING FACTORS Table 2.4 ( a and b) shows the calculated crustal and subcrustal stretching factors for the diff erent stretching events as interpreted from the subsidence cu rves for each stratigraphic column. The forward-modelled tectonic subsidence curves ( Fig. 2.27) show a remarkably good fit with the subsidence data, demonstrating that the minimisa tion technique is highly efficient. The calculated lithosphere crustal and subcrustal stretching factors reflect the accuracy level of the database and the assumptions of the model. Figure 2.27 al so shows that better age determinations for the Cretaceous sedimentary record are r eflected in better model results as compared with the Triassic and Jurassic. The lithosphere, crustal and subcrustal stretching fa ctors calculated for each finite stretching phase from forward modelling have been plotted in map view (Figs. 2.28 to 2.34). 6.3.1. M esozoic l ithosphere str etchin g phases
The lithosphere extension factors for each stretching phase are: Triassic stretching phase. Uniform stretching factors β = δ reach values up to 1.23 in the Ciénega de Morrocoya l, 1.17 in the area of the Upper MV and 1.13 in the western flank of the palaeo-EC (Fig. 2.28). Spatial distribution of values confirms small narrow (150 km wide) rift basins with opposite polarity located on opposite sides of the palaeo-Magdalena-La Salina fault sy tems. Triassic to Middle Jurassic stretching phase. The map distribution of stretching va lues (Fig. 2.29) corroborates two narrow (<150 km wide) rift basins located at the current location o f Serranía de San Lucas ( Fig. 2.1) with β = δ values up to 1.09, and inferred stretching factors β = δ in the area of the Upper MV with values up to 1.12. Lithosphere stretching β = δ values in the western flank of the palaeo-EC only reach up to 1.02. The abundance of volcanic rocks in these rift basins is suggestive of a positive thermal anomaly that probably also weakened the lithosphere (as interpreted in many rift basins, Ziegler, 1994). Lithosphere thermal doming could also have produced the observed unconformities at the bottom of the syn-rift fill. Lithosphere stretching factors play a dominant role controlling the degree of upwelling and adiabatic decompression of the asthenosphere and lower lithosphere. Partial melting occurs when the upwelling material crosses the mantle solidus line (McKenzie and Bickle, 1988; Wilson, 1993). Fast subsidence rates, en échelon pattern and high level of volcanic activity are features commonly associated with oblique slip rift zones (Ziegler, 1994). The width of these basins increased compared to early Triassic, suggesting an increasing width of the thermal weakened lithosphere. .
56
Chapter 2 MEDIOS BETA DE LTA
C AS C A JA L E S - 1 B E T A D E L T A 30 0
25 0
20 0
15 0
10 0
50
0
30 0
25 0
20 0
15 0
10 0
50
0
0
0 50 0
50 0
1000
forward observed
1500 2000
forward observed
1000 1500 2000
2500
AR C AB U C O B E TA D E L T A 30 0
25 0
20 0
15 0
10 0
50
TABL AZO BE TA DE LTA 0 0
0
30 0
25 0
20 0
15 0
10 0
50
0
50 0
50 0
1000 1000
forward observed
1500
forward observed
1500 2000 2500
2000
CH IMA BETA DEL TA 30 0
25 0
20 0
15 0
QUIPILE BETA DELTA
10 0
50
0
30 0
0
0 20 0
25 0
20 0
15 0
10 0
50
0
50 0
40 0
forward observed
60 0 80 0
forward observed
1000 1500
1000 2000
1200
S U B A- 2 B E T A D E L T A
30 0
25 0
20 0
15 0
10 0
VILL ETA BETA DE LTA 50
0
0
0
30 0
50 0
50 0
1000
1000
forward observed
1500
2000
2500
2500
3000
3000
25 0
20 0
15 0
10 0
50
0
0
forward observed
30 0
0
25 0
20 0
15 0
10 0
50
0
forward observed
1000 1500 2000
3500
2500
C AQ U E Z A B E TA D E L TA 25 0
20 0
15 0
10 0
50
Q O L I N I B E TAD E L TA 0
30 0 0
50 0 1000
50
50 0
2500 3000
30 0
10 0
S AN L U I S D E G AC E N O B E T A D E L T A
1000 1500 2000
0
15 0
forward observed
YACO P I B ET A D E L TA 30 0
20 0
1500
2000
0 50 0
25 0
25 0
20 0
15 0
10 0
50
0
50 0
forward observed
1000
1500
1500
2000
2000
forward observed
Figure 2.27: Forward modelled tectonic subsidence (continuous line) and observed tectonic subsidence (dots) curves (in meters).
57
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Q E L C O B R E B E TA D E L T A 30 0 0 20 0 40 0 60 0 80 0 1000 1200 1400 1600
25 0
20 0
15 0
10 0
50
C O E L L O B E TA D E L T A 30 0 0
0
25 0
20 0
10 0
50
0
20 0 40 0
forward observed
forward observed
60 0 80 0 1000 1200
NEIVA BETA DE LTA 30 0 0
15 0
25 0
20 0
15 0
10 0
50
PRADO BETA DE LTA 0
20 0 40 0
0
80 0 1000
25 0
20 0
15 0
10 0 20 0 30 0 40 0
forward observed
60 0
30 0
10 0
50
0
forward observed
50 0
1200
60 0
1400
70 0
ALPU JARR A BE TA DEL TA 30 0 0 20 0
25 0
20 0
15 0
10 0
50
0
40 0 60 0 80 0 1000 1200
forward observed
1400
Figure 2.27: Continued
distribution of stretching β = δ factor value s (Fig. 2.30) indicates rift basins located along the present day western flank of the EC, with β = δ val ues up to 1.39. Moreover, assuming that the thickness of the post-rift Cretaceous sections was large along the western flank of the EC northwest of Bogota, part of the Jurassic-Cretaceous subsidence is likely to be related to the evolution of this early thermal event. This consideration suggests that these basins extended southwards into the Cundinamarca region. Berriasian to Hauterivian stretching phase. Distribution of cr ustal δ stretching factor s ( Fig. 2.31) corresponds very well to the subsidence patte rns (Fig. 2.18) and isopach maps (Figs. 2.13, 2.15). They confirm the presence of a (>180 km) wide asymmetrical half-rift basin divi ded by the Santander-Floresta horst block. Maximum tectonic subsidence and crustal stretching up to δ 1.66 w as associated with the pre-Guaicáramo master normal fault system at its eastern boundar y (Fig. 2.31). A minor second order half-graben was located in the current location of th e western flank of the EC with crustal stretching values up to 1.45 ( Fig. 2.31). Small mafic intrusions described by Fabre and Delaloye (1983) coincide with areas of thin crust (crustal stretching fac tors > 1.4) and also with places of maximum stretching of the su bcrustal lithosphere ( Fig. 2.31). As a consequence of the depth dependent lithosphere rheology assumed by the model, r esults suggest that more intense stretching affected the subcrustal mantle lithosphere ( Fig. 2.32). Differences between crustal and subcrustal stretching factors suggest some decoupling occurred be tween the crust and the subcrustal lithosphere, or that an increased thermal thinning affected the mantle lithosphere. The last interpretation implies a considerable thermal anomaly produced by mantle lithosphere thinning, which seems to be supported by the presence of magmatic mafic intrusions. Middle Jurassic stretching phase. The
58
Chapter 2 1.23
Serrania de Perija
1.1
Cienaga Morrocoyal
1400
s c a u L n S a e d i a n a r r S e
A
1300
1200
Medellin
1100
B t l u t a l f Bucaramanga u o f a e a o 1.11 l e a p l a a z r e a p a c u 1.13 a S y o B
1 1 .
1000
900
e g u I b a
ra e n t C
800
800
Arauca
Llanos Orientales Basin (LLA)
Yopal
1.00
l t f a u
m t e Ibague s y s t 1.06 l 1.16 u f a o e 1.17 l a a p a 1.09 m s 1.16 u h C l 1
A Bogota 1.00
B
Villavicencio 1.00
0
Eastern part of Chibcha Terrane (Toussaint, 1995) and Guyana Shield.
LITHOSPHERE STRETCHING FACTOR TRIASSIC (248.2 - 235)
200Km
100
1.08 a e n e d r a i a n c a a r M r e a S L
900
Payande San Lucas Terranes (Etayo-Serna et al., 1986). Western parat of Chibcha Terrane (Toussaint, 1995).
1.00
1 .
Neiva
Barinas Basin
1.00
1
Manizales
s d e n A a d i r M e
Cucuta
m t e 1.00 s y 1.00 s t l u 1.05 f a 1.04 o e l a a p Eastern a Tunja m Cordillera i u t i B 1 .
ra l le i rd o C
Maracaibo Basin
1000
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.28: Contour map of lithosphere stretching factors ( β =δ) calculated through forward modelling for the Triassic (248.2-235 Ma) stretching event. assuming the hypothesis that there are Triassic sediments in the Ciénaga de Morrocoyal area (Geyer, 1982) and that Triassic sediments were accumulated in the western flank of the EC: Distribution of main early Mesozoic faults is also shown.
During rifting, stress-induced lithosphere thinning causes adiabatic decompression of the lower lithosphere and asthenosphere, their partial melting and the diapiric rise of melts into the zone of thinned lithosphere. Mafic melts appear to be generally derived from an incompatible elementenriched mantle source residing primarily in the subcrustal lithosphere and/or within mantle plumes, and from the underlying depleted asthenosphere (Wilson, 1989 in Ziegler, 1994; Wilson, 1993). Although the 1D model can not predict regional isostatic effects, the Lower Cretaceous unconformity on the rift margins ( e.g. LLA) and locally on horst blocks ( e.g. Santander-Floresta palaeo-Massif) was probably produced by thermal uplift of rift shoulders as suggested by the subcrustal stretching values. According to Ziegler (1988), Kusznir et al . (1991), Kusznir and Ziegler (1992) and Ziegler (1994), unconformities on rift shoulders and intra-basinal fault blocks can be attributed to footwall uplift in response to extensional unloading of the lithosphere. This phenomenon may be enhanced by thermal uplift of the rift zone and a gradual strain concentration in the axial rift zone. In general terms, the location of subcrustal and crustal stretched zones coincides, as a consequence of the 1D model assumption of local isostasy. However, where there is some offset, this is indicative of some asymmetry in the basin, as also indicated by the general geometry of the basin.
59
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes
Based on subsidence analysis of the Cretaceous on stratigraphic columns of the EC Hébrard (1985) and Fabre (1987), using the instantaneous stretching model of McKenzie (1978), have calculated uniform β = δ stretching factors up to 2 for the whole lithosphere. 1.09
Cienaga Morrocoyal
1400
1 . 0 5
s c a u L a n S d e i a 1.06 a n r r S e
A
1300
m te ra s y le s i l rd lt o u fa C Medellin o e la a p a im it u B
1200
1100
Maracaibo Basin
Serrania de Perija
s d e n A a d i r M e
Cucuta
B t l u f a Bucaramanga e o l a a l 1.00 t p u z f a r e o a u a e S l 1.00 a p a c 1.00 a y o 1.02 B
Arauca
Llanos Orientales Basin (LLA)
1.00 1.00
Eastern Cordillera
Tunja Yopal
A
Manizales 1.00
t a u l e f u g I b a
1000
1.00 Ibague 1.00
Bogota 1.00
. 5 1 0
1.04
1 1 .
800
800
Eastern part of Chibcha Terrane (Toussaint, 1995) and Guyana Shield.
1.06
1.00 1.12
a l n tr e C
B
Payande San Lucas Terranes (Etayo-Serna et al., 1986). Western parat of Chibcha Terrane (Toussaint, 1995).
Villavicencio
1.04
900
Barinas Basin
0
100
LITHOSPHERE STRETCHING FACTOR EARLY JURASSIC (208-185)
200Km
1.00 Neiva a e n d e r a a c i n a a r M r e a S L
900
1000
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.29: Contour map of lithosphere stretching factors ( β =δ) calculated through forward modelling for the Early Jurassic (208-185 Ma) stretching event. Distribution of main early Mesozoic faults is also shown.
However these authors lumped the Cretaceous stretching events in a single “instantaneous” stretching event with infinite extension rate. The higher stretching values obtained by these authors is the logical consequence of lumping several stretching events with finite extension rates in a single “instantaneous” event with an infinite extension rate. Aptian to Early Albian stretching phase. Th e map view distribution of stretching values (Figs. 2.33 and 2.34) and thickness maps ( Fig. 2.17) indicates that during Barremian and Aptian time extension took place in the south, in the a rea of the present day UMV. Crustal stretching factors up to 1.4 are associated with the southern segment of the Cambáo fault system and up to 1.2 in the Upper MV (Fig. 2.33). As a consequence of the depth dependent rheology assumed by the model, results suggest that str etching affected more strongly the subcrustal mantle lithosphere. Subcrust al stretching values reach up to 3.24 at the southern western flank of the EC and up to 1.6 at the UMV Fig. 2.34).
60
Chapter 2
Differences between crustal and subcrustal stretching values suggest some decoupling between crust and subcrustal lithosphere or that an increased thermal thinning affected the mantle lithosphere. These results would imply a thermal anomaly that probably is responsible of rift shoul der uplift, as interpreted from fission track data by Van der Wiel (1991) in the UMV and Garzón Mass if (Fig. 2.1). Isostatic adjustment of the crust in response to stretching compensating the thermal upli ft within the rift basin explains the absence of thermal uplift unconformities within the basin. 1.00
Cienaga Morrocoyal
1400
A
s c a u L n a S d e i a n a r r S e
1300
1200
Medellin
ra l le i rd o C
1100
Maracaibo Basin
Serrania de Perija
B 1.1
t l t l u u f a a f o o e e l a l a a a p p z a e r c a a y u S 1.00 B o
1.39
1.35
1000
1.00
900
800
m t e Ibague s 1.00 y s t l u 1.00 1.00 f a o e 1.00 1.00 l a 1.00 a p a m s u l 1.00 h C t ra
e n C
Neiva
800
1.00
1.07
?
Llanos Orientales Basin (LLA)
Tunja Yopal 1.00
?
A
Bogota
Eastern 1.00 Cordillera
B Villavicencio
0
a e n d e r a a i n c a a M r r e a S L
900
Arauca
1.09
?
l t f a u g u e I b a
Barinas Basin
Bucaramanga
1.00
m t e s y s t l u f a o e ? l a a p a m i u t i B
Manizales
e s d A n a i d r e M
Cucuta
1000
100
1200
Eastern part of Chibcha Terrane (Toussaint, 1995) and Guyana Shield.
LITHOSPHERE STRETCHING FACTOR JURASSIC (180.1 - 176)
200Km
San Jose del Guaviare
1100
Payande San Lucas Terranes (Etayo-Serna et al., 1986). Western parat of Chibcha Terrane (Toussaint, 1995).
1300
1400
1500
Figure 2.30: Contour map of lithosphere stretching factors ( β =δ) calculated through forward modelling for the Jurassic (180.1-176 Ma) stretching event. Distribution of main Early Mesozoic faults is also shown. 6.3.2. Total str etchi ng
Figure 2.35 sh ows a map of crustal thickness at the end of Cretaceous calculated using the total amount of lithosphere stretching during Mesozoic time and assuming an initial crustal thickness of 35 km. This value has been considered as representative of the undisturbed crustal initial rifting stages, thickness in the LA area (Calvache and Muñoz, 1984). No major extensional or compresional tectonic event has affected the LLA lithosphere during Mesozoic and Cenozoic time. This map represents the cumulative crustal stretching of the whole Mesozoic extensional history of the EC, MV and LLA areas.
61
Mesozoic Rifting History of the Eastern Cordillera, Colombian Andes Serrani a de Perija
1400
1300 1.08
s a c u L n a S e d Bucaramanga i a 1.21 n a 1.10 1.05 r r a e r 1.15 e l i S 1.12
d o r C 1200
Maracaib o Basin Cucuta 1.1
1.28
1.13
Arauca 1.00 1.08
9
1.48
2 1.18
7
8
1.15 1.04 1.35 1.19
1.25
6 14 15 A
13
1.02
1.04
1.10 1.23
1.00 1.33 1.21
Tunja
1.25
1.22
1.23 1.20
1.40
Ibague
1.00 1.00
17
1.00 1.00 1.00
1.00
1.00
1.00
1.00 1.20
1.00
1.00
1.00
1.00
Bogota
1.00
1.00
1.12
1.15
1.00
Yopal
1.00
1.16 1.13
Llanos Orientales Basin (LLA)
1.00
4
1.35
1.21
16
1.00
1.00
3
5
B
1.09 1.15
1.00
1.00
1.22 1.01
1.00
1.31
1.18
Manizales
1
1.18 1.30
1.21
1.00
1.00
1.00
1.00
Villavicencio
1.00
1.00
CRUSTAL STRETCHING FACTOR BERRIASIAN HAUTERIVIAN (144-127)
1.00
1.00 1.00
1.00
1.00
Neiva
1.00 0
100
1.00
200 Km
Cretaceous mafic igneous intrusions
800 1.00
Emerald mines
900
1000
1.00 1.00
1.00
1.00
1.00
1.00
1.00
800
1.00
1.25
10 to 12
900
1.00
1.35
1.14 1.26
1.24
l a r t e n 1.00 C
1.00
1.23
1.31
1000
Basin
1.66
Medellin
1100
Barinas
1.2
1.09
1.11
a r i d e M
s d e n A
1100
1200
1300
1400
1500
Figure 2.31: Contour map of crustal ( δ) lithosphere stretching factors calculated through forward modelling for the Berriasian-Hauterivian (144-127 Ma, Cretaceous) stretching event. Distribution of main Early Cretaceous faults and mafic intrusions is also shown with circles: 1. Diorite Rio Nuevo, 2. Microgabro Rodrigoque, 3. Porfiritic basaltic lava, 4. Microgabro Rio Cravo Sur, 5. Pajarito, 6. Q. La Esperanza, 7. Q. Las Palomas, 8. Q. La Culebra, 9. Marfil, 10. Q. Grande, 11. Q. La Chorrera, 12. La Chunchalita, 13. Q. La Fiebre, 14. Caceres, 15. La Corona, 16. Pacho, 17. Diorite Rio Guacavia.
7. DISCUSSION 7.1. GEOMETRY OF RIFT BASINS Triassic rift basins were narrow and increased in width and surface area during Triassic and Jurassic times. Cretaceous rifts were wider, covered more area and were more asymmetrical than Triassic-Jurassic rift basins. In many are as it has been observed that during reactivation of
62
Chapter 2
crustal discontinuities can lead to the subsidence of isolated grabens and half grabens that are linked by shear zones ( e.g. east African rift). With increasing strain, such grabens propagate toward each other, coalesce and evolve into a more or less continuous rift system (Nelson et al., 1992; Ziegler, 1994). According to Ziegler (1994), propagation of established rift systems into previously unextended areas could occur either by reactivation of pre-existing crustal discontinuities, or through the development of new fault systems crosscutting the basement grain. Maracaibo Basin
Serrania de Perija
s ca u L n a S e d Bucaramanga ia n 1.05 1.37 1.31 a 1.52 rr e 1.30 S
1400
1300 1.08
1200
Medellin
r a l le i r d o C
. 2 1
Cucuta
1.63
1.72
1. 1 1. 2 1. 3 1 .4
1 . 6
5 1 .
1.00
e m t s 1.64 s y 3.49 0 t 0 . l 3 . 2 u f a 2.53 o a e l p a 1.00 o m a r 1.00 c á i u a 1.00 G
m t e s 1.35 y . 3 p a 1.59 t s 4 l 1 . a u 1.38 1 c f a 1 y a . 2 1.02 o e o B 1.04 a 1 l a 1.10 1 . 3 2.05 2.07 p . 1 a 2.58 m 1.45T r a 1.25 1.23 Tunja i n a u t i 2.20 3 . s f 0 B e r 1.21 1.21 2 .0 p a 1. 1 l a 1 e o . 1.00 1 1.03 f 1.22 1.01 a u l t 1.08 1.23 1.00
0 2.
1100
. 1 1
2.02
1.00 3.15
1.17
1.02
Manizales
1.09
1000
1 1 1.79 . 7 . 6
1.00
Ibague 1.00
1.57
1.00
1.00 1.00 1.00
900
1.00
l r a t e n C
1.1
1.33
1 1 Bogota
N . 5 . 4 1.22 a z a r 1 . e t 3 h p a l a e o f a u 1.00 l t
Yopal
1.00
1.00
1 .2 1 1 .
1.00 1.00 1.00
1.00
1.00
1.00
1.00
1.00
Villavicencio
e a d n e a r a i n c a a r M r e a S L
1000
1.00
1.00 1.00
0
SUBCRUSTAL STRETCHING FACTOR BERRIASIAN HAUTERIVIAN (144-127)
200Km
100
1.00
900
Llanos Orientales Basin (LLA)
1.00
1.00
800
1.00
1.00
1.00 1.00
1.00
1.00
1.00
1.00
3.61
1.00
1.00
Neiva
Arauca
1.00
1.00
1.00
800
basin
1.36
1.15
2.70
Barinas
1. 5
t l u 1.26 f a e o 1.13 a l a 1 l t 1.57 p 1. z a u 3 f e . r 1.25 o 2 1 u a 1.47 e . S l a 1
1.06
. 1 1
e s d A n a i d r e M
1.00
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.32: Contour map of subcrustal ( β ) lithosphere stretching factors calculated through forward modelling for the Berriasian-Hauterivian (144-127 Ma, Cretaceous) stretching event. Distribution of main Early Cretaceous faults is also shown.
The evolution from asymmetrical Triassic-Jurassic rifts to more asymmetrical Cretaceous rifts may suggest a transition from a more pure shear rifting during Triassic to a simple shear rifting during Cretaceous. However Radelli (1967, in Toussaint, 1995b) has pointed out eastward tilted faulted blocks during Triassic and Jurassic times implying that asymmetrical basins existed. Whether the Colombian Mesozoic extensional basins were pure shear rifts or simple shear rifts is difficult to demonstrate. Probably both mechanisms were operating; these rift models should be viewed as end member cases. Lithosphere extension is localized in zones where the lithosphere is weakest. Such zones correspond to areas of thermal destabilized lithosphere such as areas of previous rifting or in areas of crustal thickening in orogenic belts (Ziegler, 1994).
63
Mesozoic Rifting History of the Eastern Easte rn Cordillera, Colombian Andes s a c u L n a S e d ia n a rr e 1.04 S
1400
1300
Maracaibo Basin
Serrania de Perija
1.00
1.00
Bucaramanga
1.00
1.00
ra e il l rd o C
Medellin
1.00
1100
1.11 1.00 1.09
m t e s y s t l u f a o e l a a 1.30 p a i m u t i B
1.00
Manizales
1.49
1.12
. 1 1
Ibague
1.01
1.01 1.02 1.11 1.08
1.01
1.00 1.20 1.32
1.00
1.00
1.04
1.00
1.00 T r a n 1.00 T s f e r a n s 1.18 r f a f u l t e r p a 1.25 l a e 1.00 o 1.2 f 1.16 a u 1.14 l t 1.00
m t e s y s t l u f a o e l a a p o á m i c a u G 1.00
1.00
1.00
1.00 1.02
Yopal
1.00
1.00 1.00
1.00 1.00
1.00
1.00 1.00
1.00
1.00
1.00
1.00 1.00
1.00
1.00
1.00
1.00
1.00 1.00
1.00
Villavicencio
1.00
1.00
1.00
1.00
0
100
CRUSTAL STRETCHING FACTOR APTIAN (121-102.6)
200 Km 1.00
1.08
800
900
1.00
1.00 1.00
1.09
1.00
1.00
1.00
1.22
Neiva
1.00
1.00
1.00
800
Llanos Orientales Basin (LLA)
1.00
1.00
1.12
a l n tr e C
1.00
1.00
1.10 1.05
1.00
1.00
1.00
1.00
Arauca Arauca
1.00
1.00
Tunja
1.00
1.14
1.00
1.00
1.00
1.00
1.05
1.00
1.00
1.00
1.28
900
t l u a1.00 f o e a 1.00 l a p a t i v r e S 1.00
1.07
N a 1.17 1.05 z a Bogota r e 1.00 t h 1.07 1.07 1.16 p a 1.02 l a e 1.12 1.00 o f a u l t
1000
Barinas Basin
1.00 1.13
1.03
1200
s d e n A a d i r M e
Cucuta
e a d n e r a a i c n a a r r M e a S L
1000
1.00
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.33: Contour map of crustal ( δ) lithosphere stretching factors calculated through forward modelling for the Aptian (121-102.6 Ma, Cretaceous) stretching event. Distribution of main Early Cretaceous faults is also shown.
If the orientation of pre-existing crustal discontinuities is such that they cannot be reactivated by the stress system governing the evolution of the rift, new faults will develop and “pure shear” deformation is likely to prevail (Ziegler, 1990; 1994). This mechanism may be applicable to the Triassic-Jurassic Triassic-Jura ssic rift system in Colombia. In contrast, if the upper crust is weakened by the presence of pre-existing crustal discontinuities with a favourable orientation to be reactivated under the prevailing revailing tensional stress field, these will present zones of preferential preferential strain concentration, concentration, even if they are located at considerable distances to the sides of the zone of mantle lithosphere stretching (Ziegler, 1994). This can result in “simple shear” deformation (Sawyer and Harry, 1991, Harry and Sawyer, 1992; Ziegler, 1994). This mechanism may explain the development of the palaeo-EC rift system during the Early Cretaceous. The eastern side of the rift probably developed during Berriasian-Hauterivian time by reactivation of an older Palaeozoic rift system along the Guaicáramo palaeo-fault alaeo-fault ( c.f. Hossack et al., 1999). The western side of earlier normal fault systems developed during Triassic-Jurassic rifting.
64
Chapter 2
The rheological properties of the lithosphere control the depth at which tensional necking occurs and whether a rift zone is flexed upwards or downward (Braun and Beaumont, 1989; Ziegler, 1994). A deep lithosphere necking level causes upward flexure of the rift zone. Necking at shallow crustal levels causes downward flexure of the rift zone and absence of shoulder uplifts (Kooi, 1991; Kooi et al., 1992; Ziegler, 1994). The narrow rifting during Triassic-Jurassic as well as the presence of unconformities suggest upward flexure and generation of rift shoulders, probably associated with a deep level of necking. Similar deep levels of necking in the eastern side of the Early Cretaceous rift system may have generated shoulder uplift in the LLA area during Early Cretaceous. Coarse detrital fragments in the Lower Cretaceous Brechas de Buenavista Fm. (Pimpirev et al., 1992) and Calizas del Guavio Fm. (Conglomerado de Miralindo, Ullóa and Rodríguez, 1976a) could be derived from this rift shoulder. In contrast, in the western margin of the Early Cretaceous rift system, sedimentation was more continuous from Jurassic to Early Cretaceous time implying downward flexure of the rift shoulders and thus, a shallower level of necking in this western part of the basin during Early Cretaceous times. In the newly rifted areas (western side during Triassic-Jurassic and eastern side during Early Cretaceous) necking level was deep, whereas in the previously rifted areas (western side during Early Cretaceous) necking level was shallow. Therefore, necking started at a deep level during the first stages of rifting, it then evolved to shallow-necking levels for the more mature rifting stages. Ziegler (1994) explained that rifting, involving mechanical and thermal thinning of the lithosphere, is accompanied by a gradual rise in lithospheric isotherms which entails an upward shift of the lithospheric necking level as well as of the intracrustal brittle/ductile deformation boundary. According to model experiments, the width of the rift zone depends on the thickness of the crust and the depth at which the brittle/ductile transition zone is located at the onset of the lithosphere extension (Allemand and Brun, 1991; Buck, 1991). As this interface rises with rising isotherms, upper crustal strain concentrates in time on a narrower zone (Sawyer and Harry, 1991; Ziegler, 1994). The observations in the Colombian case are opposite to these modelling results. Probably the increasing width of the Colombian rift system was the result of progressive tensional reactivation of old upper crustal weakness zones, which was not considered in those experiments. On a lithospheric scale the location of rift systems is controlled by the location of weakness zones in the lithosphere, which in turn depends on its thermal state and the thickness of the crust. At crustal scales the composition, thickness of its mechanically strong upper layer and the availability of internal discontinuities which can tensionally be reactivated, are also important controls for the location of rifts (Ziegler, 1994). The overall pattern of these rift basins for most of Mesozoic time indicates several rifts NNE-SSW oriented in an en-échelon pattern compared to the more N-S oriented Central Cordillera ( e.g. Mojica et al., 1996). Some authors (Fabre 1987; Sarmiento, 1989; Geotec, 1992; Mojica et al. 1996) have suggested also that some NW -SE faults probably probably represented transfer faults. Some features, such as the Nazareth NW-SE Fault ( Fig. 2.31), limiting limiting the Early Cretaceous basin in in the south (F abre 1987) or the NW-SE alignment alignm ent connecting the two emerald districts of the EC (Fig. 2.31, Sarmiento 1989), probably represent Mesozoic transfer faults. If a subduction-related magmatic ar c existed at the current location of the Central Cordillera during Triassic-Jurassic and possibly (?) Cretaceous times, as has been proposed by many authors (Barrero and Vesga, 1976 in Toussaint, 1995b; Aspden et al., 1987; Toussaint and Restrepo, 1974a, 1989, 1994), and if the orientation of that magmatic arc and the rifted basins has been preserved, the oblique orientation of these rifts and its en-échelon pattern would suggest some oblique slip extension with a left-lateral strike-slip component. However, with the available data it is not possible to rule-out the hypothesis that some rift arms form acute angles to the dominant NNE-SSW trend in a pattern similar similar to aborted aulacogen rifts.
65
Mesozoic Rifting History of the Eastern Cordillera, C ordillera, Colombian Andes
For example Mojica et al., (1996) have hypothesized the existence of NW-SE rift segments forming oblique angles to the dominant NNE-SSW rift segments. Branching of rift systems is of common occurrence. Most rift basins consist of half-graben depressions, the polarity of which often changes along trend across accommodation transfer zones (Ziegler, 1994). The latter are characterized by a complex fault geometry involving local positive and negative flower structures and folding (Rosendahl, 1987; Morley et al., 1990; Ziegler, 1994). In plan view, master faults of half grabens often display a curvilinear geometry. This is indicative of their listric configuration whereby their detachment level corresponds to the crustal brittle-ductile transition zone below which deformation is dominated by ductile shear (Gibs, 1987, 1989, in Ziegler, 1994). Deviations from such an idealized fault geometry can be generally related to non-orthogonal extension (Avraham, 1992) or to the reactivation of pre- existing crustal discontinuities. However, planar faults in rift zones have also been observed. observed. Mixed planar and listric fault geometries geometries are typical for many rifts (Ziegler, 1994). It is very difficult to reconstruct the geometry of the Mesozoic extensional faults in the study area. If the inverse or thrust faults that now define the eastern and western borders of the EC originally were normal faults that were inverted during the Cenozoic, their geometry in map view would provide some information about Mesozoic extensional faults. Lateral changes of Mesozoic thickness suggest that this is the case at least for the master faults that probably defined the regional rift geometry. Adopting this hypothesis, I have suggested that the Guaicáramo, La Salina, Bitúima, Magdalena, Boyacá, and Chusma Faults represent original extensional faults. The oblique orientation of most of them relative to the Mesozoic magmatic arc of the Central Cordillera would imply an oblique slip extension during the Mesozoic. Analog model experiments of oblique extension produce a similar map view fault pattern ( e.g. Tron and Brun, 1991). However it is important to keep in mind that during the Mesozoic, modifications in the drift pattern of plates, and consequently their interaction, may have resulted in changes in the stress regime governing the subsidence and structural style or development of branches of the rift system. Such interactions can result in transitions from orthogonal to oblique extension or even to wrench deformation (Ziegler, 1994). NW-SE transfer faults and possible NW-SE normal faults, as those interpreted by Ecopetrol et al . (1994) in the Middle MV, were not inverted during the Cenozoic. Delineation of basin geometry for Triassic and Jurassic times strongly depends on the assumptions on age, in the absence of accurate age data. For example for the Early Triassic, Mojica et al. (1996) considered that a single rift basin was located in the area of the Upper MV. This is based on the the idea that the Luisa Luisa Fm is the only only early Triassic sedimentary sedimentary record record in the whole whole area. If one accepts the hypothesis of Geyer (1982), based on facies and stratigraphic position, that correlative early Triassic deposits exist in the Ciénaga de Morrocoyal area, a second rift basin can be postulated ostulated in that area. In general for the Triassic and Jurassic a scenario of at least three separated rift systems (one in the Upper MV, a second in the area of the western flank of the EC, and a third in the Serranía de San Lucas) requires testing with new data. Mojica et al . (1996) have proposed several branches for the the early Mesozoic Mesozoic rift system, while Geotec Geotec (1992) and Cediel Cediel et al . (1997) proposed a large number of small rift basins. An equally acceptable hypothesis is to assume a single rift system. Such discrepancy in interpreting the geometry of rift basins is the consequence of a limited number of outcrops/well data for Lower Mesozoic sedimentary record. However, all the different hypotheses on the development of early Mesozoic rifting have in common postulated rift basins with an orientation NNE-SSW.
7.2. RELATIONSHIPS BETWEEN MESOZOIC RIFTING AND MAGMATISM Two alternative hypotheses have been proposed for the geotectonic setting of Jurassic volcanism in Colombia:
66
Chapter 2 Maracaibo Basin
Serrania de Perija
1400
s a c u L n a S e d i a n a rr e S
1300 1.00
ra e il l rd o C
1200
1.00
1.00
Manizales
1.33
t l u a f 1.00 o e a 1.00 l a p á t i v r e S 1.00
1.00
1.26
1.00
m t e t s l y u s t f a l u o f a e o a l e a l a p p a z o r e a m a r u á S i c u a G
l t 1.00
1 3.
Arauca
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Llanos Orientales Basin (LLA)
1.00
1.00 1.00 1.00
1.00
1.00
1.00
1.00
1.00
Bogota
1.00
1.00
1.00
1.00 1.00 1.00
1.47
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Villavicencio
1.00
1.00
1.00
1.00 1.00
1.00
1.00 1.00
1.42
1.14
4 1 . 2 . . 1 1 3
1.00
1.00
1.00
. 1 1
1.00
1.00
200Km
100 100
0
SUBCRUSTAL STRETCHING FACTOR APTIAN (121-102.6)
1.00 1.31
e a d n e r a a i n c a a r M r e a S L
Neiva 1.18
800
1.01
1.06
1.48 1.17
800
1.00
1 . 1
N a z a r 1.00 e t h p a 3 1.18 1. 3 l a e o f a u 1.00 l t
1.66
a l n tr e C
Bucaramanga
1.00
1.06
1.11 1.37 1.27
900
1.00
1.00
1.03
1.58 1.22
Barinas Basin
1.00
1.48
1.04
1 1 .
1.37
Ibague
2 . 1
3 . 1.13 1
1.30
1000
Cucuta
m 1.33 t e s y 1.37 1.2 1.2 s 1.00 1.22 1 . t 1.00 l 1 1.00 1.07 1 .5 u 1.10 1.00 1 1.00 .7 f a 1.00 1.05 1.72 o 3.0 Tunja e 3.24 1.77 l a 1.00 a 1.00 Yopal 1.00 p T r a T r 1.00 a n n s a s f f e r e r m p a i f a u 1.10 1.00 u l a e 1.00 l t t 2.0 2.0 i 1.52 o 1.00 1.00 B f 1.48 a 1.42 u
Medellin
1100
e s d A n a i d r e M
900
1000
1.00
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.34: Contour map of subcrustal (β ) lithosphere stretching factors calculated through forward modelling for the Aptian (121-102.6 Ma, Cretaceous) stretching event. Distribution of main Early Cretaceous faults is also shown.
(1) A subduction-related subduction-r elated magmatic arc (Toussaint and Restrepo, 1974a; Núñez, 1978; Barrero, 1979 among others). Jurassic volcanism was alkaline and related to the development of a pericratonic ericratonic arc, implying a specific coupling coupling between the South American and Pacific plates (Restrepo-Pace, 1995). (2) Rifting processes related to lithosphere extension (Estrada, 1972; Macía and Mojica, 1981; Cediel, 1983; Mojica and Macía, 1992; Macía, 1995; Mojica et al .,., 1996). Bayona et al . (1994) based on geochemic geochemical al data suggested suggested back-arc back-arc volcanism. volcanism. In the study area in contrast to the alkaline Jurassic volcanism nature of Cretaceous magmatism is different. During the Cretaceous there was still some alkaline volcanism (tuffs and bentonites, Villamil, 1994; Restrepo-Pace, personal communication), but also intrusion of diabasic-gabroic dikes with a toleithic affinity (Fabre and Delaloye, 1983; Moreno and Concha, 1993). The latter probably more related to extension\transtension. The following discussion is based on the assumption that Mesozoic volcanic rocks in the Mesozoic extensional basins would be related to rifting processes as suggested by Mojica et al . (1996).
67
Mesozoic Rifting History of the Eastern Easte rn Cordillera, Colombian Andes 3 0
a s c L u n a S e d a i a n r r e 28,184 S 29,200
1400
1300 ra i l le rd o 3 0 C 1200
2 5
Bucaramanga 31,760
19,63 7
2 5
24,048
24,095
30
28,294
1100
34,044
16,869
Manizales
21,625
24,179
31,818
24,386
29,536 23,632
22,508
1000
29,838
Ibague 27,649
3 5
24,352
Yopal Yo pal 34,014 35
23,680
35
24,331
35
Llanos Orientales Basin (LLA)
29,518
24,52 24, 909 8 24,909 24,636
22,439
2 5
35
28,229
35
28,664
29,636
35
35
29,636 27,999
Tunja
15,853
3 0
2 5
22,115
3 0
35
21,123
22,774
20,315
2 0
35
28,690
28,398
24,977 23,117 12,164
Arauca Arauca
26,085 26,889
2 5
25,15 2
35
25,659
30,310
28,780
20
35
25
28,153
Medellin
Barinas Basin
23,460
30,514 26,132
s d e n A a d i r M e
0
Cucuta
27,344
29,651
32,37 7
Maracaibo Basin 3
Serrania de Perija
35
28,920
25,11 2
35
35
30,756
24,418
28,504
30
Villavicencio
27,259
35
35
31,818 35
21,758
900
28,928
al n tr e C
25
27,673
-
200 km
35
a e n d e r a a i c n a a r M r e a S L
30
32,051
900
1000
CRUSTAL THICKNESS AT THE END O F CRETACEOUS (Km)
35
3 5
Neiva
800
100
0
28,419
800
35
35
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 2.35: Contour map of crustal thickness (km) at the end of Cretaceous time, calculated using the total amount of stretching during the Mesozoic and assuming an initial crustal thickness of 35 km before stretching. 35 km is a representative value of the undisturbed crustal thickness in the Llanos Orientales area according to Calvache and Munoz (1984).
According to McKenzie and Bickle (1988) and Wilson (1993) the amount of lithospheric stretching plays a dominant role by controlling the degree of upwelling and adiabatic decompression of the asthenosphere and lower lithosphere. Partial melting occurs when the upwelling material crosses the mantle solidus line whose position on P-T space is a function of composition. However, in the study area abundant Late Triassic-Early Jurassic volcanic rocks are associated with moderate stretching factors ( β = δ up to 1.12) while the Cretaceous sedimentary record is almost devoid of volcanic rocks (only containing minor mafic intrusions), is associated to higher stretching factors ( β up to 3, δ up to 1.66). Clearly thermal processes were more important compared to mechanical stretching during Late Triassic-Early Jurassic rifting than during Cretaceous rifting. During the Late Triassic-Early Jurassic, abundant volcanic rocks suggest a positive thermal anomaly in the lithosphere but a moderate lithosphere stretching. Triassic-Jurassic unconformities could have been produced by thermal uplift uplift (“active rifting”?). rifting”?). Thermal doming is the result of progressive thinning of the higher density mantle lithosphere and its replacement by low-density asthenosphere (Bott, 1992). In contrast during Cretaceous time
68
Chapter 2
the much less abundant volcanic rocks, absence of tectonically controlled unconformities, and the large amount of tectonic subsidence suggest absence of thermal doming. The presence of small mafic intrusions coinciding with places of maximum crustal and mantle subcrustal stretching suggests modest magmatism as a consequence of extension of the lithosphere (“passive rifting”). Subsidence in rift basins is a consequence of isostatic adjustment of the crust crust to lithospheric lithospheri c stretching. It tends to be counteracted by uplift induced by thermal and mechanical attenuation of the subcrustal lithosphere (Ziegler, 1994). Subsidence patterns and the relative abundance of volcanic rocks suggest that during Late Triassic-Early Jurassic times thermal controlled uplift counteracting subsidence was more important than during Cretaceous times. The plutonic bodies of the Santander Massif, radiometrically dated as Palaeozoic (Goldsmith et al .,., 1971; Boinet et al . , , 1985b; Restrepo-Pace, 1995) or Triassic-Jurassic (Goldsmith et al .,., 1971; Ward et al .,., 1973; Restrepo-Pace, 1995), could represent Palaeozoic bodies reheated during Jurassic time, as well as early Mesozoic intrusions. The thermal anomaly that affected them during Jurassic also may also have generated some degree of thermal uplift that counteracted subsidence in the less subsiding Santander-Floresta Massif high block. Mesozoic rifting in the study area implies that tensional stresses affected the lithosphere. However probably shear-traction exerted by the upward and outward asthenospheric flow enhanced these stresses during Triassic-Jurassic times to the degree that crustal discontinuities could have been tensionally reactivated. In contrast during Cretaceous times the asthenospheric effect was probably less important and the development of a thermal anomaly and magmatism was mainly favoured by mechanical stretching in a more “passive” mode. Wrench induced pull-apart basins and oblique slip rift zones often display a relatively high level of volcanic activity (Ziegler, 1994). Wilson and Guiraud (1992) and Ziegler (1994) suggested that major wrench faults transect the entire lithosphere, thus providing conduits for magma migration to the surface. If this concept is applicable to the Colombian case, the hypothesis of a single narrow rift system can be regarded as a broad wrench zone. The hypothesis of several small narrow rift systems with en- échelon pattern may also support this idea. Toussaint (1995b) reported some diabase dikes that were controlled by small pull apart structures related to the left-lateral strike-slip normal faults in the Early Jurassic Saldaña Fm of the Upper MV, which supports this idea. Volcanic rocks associated with intra-continental rifts display a typically alkaline, mafic/felsic bi-modal composition (Wilson 1989, in Ziegler, 1994; Wilson, 1993). The abundant Late TriassicEarly Jurassic volcanic rocks of the study area have variable composition from felsic to mafic. Chemical analyses of La Quinta Fm. volcanic rocks indicate calc-alkaline composition in the diagram AFM and alkaline composition in the alkali-silica diagram (Toussaint, 1995b). Chemical analyses of the Saldaña Fm indicate calc-alkaline composition probably generated in a backarc environment (Bayona et al., 1994). It is generally accepted that magmatic arcs of calc-alkaline composition located on continental borders are related to subduction (Aspden et al., 1987). The predominance redominance of calc-alkaline calc-alkaline composition, composition, however, seems to suggest a convergent-relate convergent-relatedd rifting rather than intra-continental intra-continental rifting (Toussaint, 1995b). 1995b). Maze (1984) based on Sr isotopic values values comprised between 0.705 and 0.715 has suggested a partial cortical source for these magmas. In many rift systems, the mafic melts appear to be generally derived from an incompatible elementenriched source, residing presumably in the subcrustal lithosphere and/or within the mantle plumes, and from the underlying depleted asthenosphere (Wilson, 1989 in Ziegler, 1994; Wilson, 1993). During the evolution of some rifts, decrease in alkalinity of the extruded mafic magmas, and an increasing contribution of mid ocean-ridge basalt (MORB) source (depleted mantle) melts can be recognized, both in time and generally toward the rift axis. This can be attributed to an increasing contribution from melts from the asthenosphere, as the lithosphere is progressively thinned (Wilson, 1993; Ziegler, 1994). These processes provide an explanation for the change from felsic composition at bottom to intermediate at top, as reported by Bayona et al. (1994) for the volcanic rocks of the Jurassic Saldaña Fm. Also these processes explain the mafic calc-alkaline up to tholeitic composition (Fabre and Delaloye 1983) of the mafic Cretaceous intrusions located in the central part of the basin where lithosphere was most thinned. The calculation of stretching factors based on subsidence analysis does not consider flexural effects, intra-plate stresses or possible crustal density changes. Stretching factors derived from deep crustal seismic or measurement of crustal extension by faulting from structural sections are generally
69
Mesozoic Rifting History of the Eastern Easte rn Cordillera, Colombian Andes
smaller than those calculated from subsidence analysis because the model does not take into account these processes (Ziegler, 1994). Nevertheless this quantitative subsidence analysis contributes substantially to the understanding of post-rift subsidence processes, providing a measure of the thermal anomaly that was introduced during the rifting stage of the basin. A clear understanding of the thermal regime of a basin is of primarily importance for modelling petroleum generation.
7.3. SUGGESTIONS FOR FUTURE STUDIES It is generally accepted that extensional basins were developed in the area of the EC and MV during the Mesozoic ( e.g. Fabre, 1983a,b, 1987; Cooper et al., 1995). One of the limitations in Colombia, especially for the continental Triassic and Jurassic poorly fossiliferous sedimentary record is the scarcity of accurate age data. In some cases deposition age has been inferred from relative stratigraphic positions with respect to fossiliferous units, or by simple lithological or facies correlation assuming facies synchroneity. Another limitation results from the controversial ages proposed by different authors for a single lithostratigraphical lithostratigraphical unit. Better data are necessary to properly quantify the early Mesozoic of the EC and MV. Further stratigraphical stratigraphical studies of the early Mesozoic subsidence history of the EC and MV are strongly recommended. Even for the Cretaceous marine sedimentary record, which is better dated by biostratigraphy, there is a need for better data. This Cretaceous record is locally composed of monotonous shale successions disarmonically folded, where thickness are difficult to measure and structural repetitions difficult to detect, in the absence of a high number of age control points (Restrepo-Pace, 1989). Better data are necessary to accurately quantify the Mesozoic subsidence. It is extremely important to understand basin subsidence during Mesozoic time to understand the Cenozoic tectonic evolution, because Mesozoic rifting weakened the lithosphere and created cre ated weakness zones that that strongly affected later deformation and basin formation processes (see (se e Chapters 3 and and 4). In addition Mesozoic rifting provided heat to start organic matter maturation maturati on and hydrocar hydroc ar bon generation generation processes, which are a key element for hydrocarbon exploration.
8. CONCLUSIONS The high-resolution backstripping analysis and forward modelling showed that Mesozoic Colombian basin is marked by five rifting pulses. Periods of rift activity correlate in time with gaps of subduction-related magmatic arc activity as suggested by Aspden et al. (1987) especially for Jurassic time, supporting the hypothesis of backarc extension. If backarc extension continued during the Early Cretaceous by oblique plate convergence, it probably has a strong strike-slip component, as suggested by Aspden et al . (1987). Evidence supporting the hypothesis of a backarc basin located behind a partially emerged less subsiding subsiding palaeo-Central palaeo-Central Cordillera (magmatic arc?) could be summarised as follows: (1) The presence, in the western part of the Cundinamarca sub-basin, of Lower Cretaceous sandstones with abundant volcanic lithic fragments and feldspar derived from a western detrital source area as indicated by palaeocurrent data (Murca Fm. And Útica Sandstone; Sarmiento, 1989; Moreno, 1990b, 1991); (2) The presence of progressive westerly onlap terminations of the Cretaceous carbonates on th e basement, observed observed in seismic lines, in the western border of the Cesar Valley, in northern Colom Colom bia (Fig. 2.1; Audemard, 1991); (3) Petrographical evidence suggesting that Berriasian (?) to V alanginian clastic clasti c sediments near San Felix in the western flank of the Central Cordillera came from erosion o f nearly uplifted areas containing fragments of metamorphic rocks and small tectonic blocks with plutonic rocks (Rodríguez and Rojas, 1985). (4) Cretaceous volaniclastic rocks (Rodríguez and Rojas, 1985; Rubiano, 1989; Villamil and Arango, 1998) that were probably also derived from a magmatic arc; (5) Late Cretaceous zircon fission track ages in the Central Cordillera (Gomez et al , 1999; Toro et al , 1999); (6) Some Cretaceous plutonic bodies in the Central Cordillera. However, the hypotheses of a passive margin (Pindell and Erikson, 1993) or aborted rift arms related to the break-up of Pangaea (Geotec, 1992) cannot be completely ruled out due to the absence of a well-defined Cretaceous magmatic arc. Three stretching events are suggested during Triassic-Jurassic time. Spatial distribution of values suggest that small narrow (<150 km wide) asymmetric rift basins were located on opposite sides of the palaeo-Magdalena-La Salina fault system, which probably was active as a master normal fault
70
Chapter 2
system. During Berriasian-Hauterivian subsidence curves and isopach maps suggest that a (>180 km) wide asymmetrical half-rift basin existed divided by the Santander Floresta high block. In the south there was a single depocenter limited southward by a vertical transfer fault. Location of small mafic intrusions (Fabre and Delaloye, 1983) coincides with areas of thin crust (crustal stretching factors > 1.4) and also with places of maximum stretching of the subcrustal lithosphere. During Aptian to early Albian time, the basin extended towards the south in the UMV. Differences between crustal and subcrustal stretching values suggest either that there was some lowermost crustal decoupling between crust and subcrustal lithosphere, or that an increased thermal thinning affected the mantle lithosphere. Triassic rift basins were narrow and increased in width during Triassic and Jurassic times. Cretaceous rifts were wider, and were more asymmetrical than Triassic-Jurassic rift basins. During Berriasian-Hauterivian time the eastern side of the rift possibly was developed by reactivation of an older Palaeozoic rift system associated to the Guaicáramo fault system ( c.f. Hossack, et al., 1999). The western side probably developed by reactivation of an earlier normal fault system developed during Triassic-Jurassic rifting. During the first stages of rifting, lithosphere necking started at a deep level, and evolved to shallower necking levels in the latter rifting stages. Probably the increasing width of the rift system was the result of progressive tensional reactivation of pre-existing upper crustal weakness zones. Lateral changes of Mesozoic sediment thickness suggest that the reverse or thrust faults that now define the eastern and western borders of the EC were originally normal faults inverted during the Cenozoic Andean orogeny. Thus the Guaicáramo, La Salina, Bitúima, Magdalena, Boyacá, and Chusma were originally extensional faults. The oblique orientation of most of them relative to the Mesozoic magmatic arc of the Central Cordillera may be the result of oblique slip extension during Mesozoic, or alternatively can be inherited from the pre-Mesozoic structural grain. However, not all the Mesozoic extensional faults were inverted ( e.g. NW-SE in the Middle MV), some normal faults were passively transported with short-cut basement blocks during Cenozoic inversion ( e.g.E smeraldas Fault, ESRI and Ecopetrol, 1994, Cooper et al .,., 1995). Thermal processes were more dominant than mechanical stretching during Late TriassicEarly Jurassic phase than during Cretaceous rifting phase. During Late Triassic-Early Jurassic abundant volcaniclastic rocks suggest a positive thermal anomaly in the lithosphere but a moderate lithosphere stretching. Triassic-Jurassic age unconformities could have been produced by thermal uplift (“active rifting”?). In contrast during the Cretaceous less abundant volcanic rocks, absence of tectonically controlled unconformities, and the large amount of tectonic subsidence indicates absence of thermal doming. The presence of minor mafic intrusions coinciding with places of maximum crustal and mantle subcrustal stretching, suggest that a modest magmatism took place as a consequence of extension of the lithosphere (“passive rifting”).
71
CHAPTER 3 PALAEOGENE INCIPIENT BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES 1. INTRODUCTION Inverted extensional basins and upthrust basement blocks are common intraplate compressional/transpressional structures associated with continental plate margins (Ziegler et al., 1998). Such compressional features at plate margins commonly result from collision-related orogenic processes. Compressional/transpres Compressional/transpressional sional stresses that are related to collisional collisional plate interaction interaction are responsible for inversion of tensional hanging-wall basins (Ziegler et al., 1995). Compression in the regions behind a magmatic arc is associated with Andean type orogens and occurs during periods of increased convergence rates between the subducting and overriding plates (Ziegler et al., 1998). Inversion of tensional hanging-wall rift basins located behind a magmatic arc is the result of acceleration of convergence rates between the colliding plates, their increased mechanical coupling and the transmission of compressional stresses into the backarc domain of the overriding plate (Uyeda and McCabe, 1983; Ziegler, 1993; Ziegler et al., 1998). The Palaeogene plate tectonic history of Colombia seems to provide an appropriate tectonic setting for inversion of Mesozoic extensional basins. Some authors ( e.g. Van der Hammen, 1961; Roeder and Chamberlain, 1995; Restrepo-Pace et al., 1999a,b; Gomez et al., 1999) have suggested the possibility ossibility that upthrust ed blocks and/or incipient inversion of Mesozoic extensional basins in the area of the EC ( Fig. 3.1) during Palaeogene created some topography. However, the most commonly accepted view view is that of of a unique simple foreland basin related to the topographic load of the Central Cordillera ( e.g. Cooper et al., 1995, Fig. 1.2). Typical palaeogeographic Typical palaeogeographic interpretations show the area east of the Central Cordillera as a s ingle basin where sedimentation covered its entire surface during early Tertiary times. Was the area east of the Central Cordillera (MV, EC and LLA) a single simple foreland basin (Cooper et al., 1995, Fig. 1.2) or a compartmentalized basin with local palaeotopo alaeotopographic graphic emergent areas resulting from incipient inversion of Mesozoic extensional basins (Fig. 3.1, Gomez et al., 1999) during Palaeogene time? The The aim of this Chapter is to constrain alternative possible tectonic scenarios for the Palaeogene tectonic history of the EC. To this purpose, local detailed geological evidence is compiled into a regional geological model and quantitative modelling of tectonic thermal and flexural subsidence is carried out to test these different tectonic scenarios. Additionally this chapter contributes to understanding of the Palaeogene tectonic basin forming and inversion mechanisms in terms of geodynamic processes that govern deformation of the lithosphere. I attempt to put the Palaeogene tectonics in the study area in a plate-tectonics framework addressing the mechanical control on the tectonic processes, such as the role of stresses, lithosphere rheology, and pre-existing crustal discontinuities. In this Chapter and subsequent chapters I discussed flexural subsidence of lithosphere produced by topography topography generated by uplift. In order to discusse uplift I use the terminology proposed proposed by England and Molnar (1990) as defined in equation (3.1). Surface uplift ( ∆H) is the upward displacement of earth’s surface with respect to the geoid. Uplift of rocks (U) is the upward displacement of the rocks with respect to the geoid. Exhumation (E) is the upward displacement of rocks with respect to the earth´s surface. Exhumation and denudation are both the same, but have opposite signs; denudation is the downward movement of earth’s surface relative to a fixed rock volume. These variables are related: (3.1) ∆ H = U – E
Palaeogene Incipient Basin Inversion Inver sion History of the Eastern Easter n Cordillera, Colombian Andes JURASSIC Palaeo-Magdalena Valley
Magdalena-Tablazo sub-basin
Santander-Floresta palaeo-high
63 km (from average Cretaceous crustal stretching factor 1.2)
Moho
CRETACEOUS Palaeo-Magdalena Valley
Magdalena-Tablazo sub-basin
Santander-Floresta palaeo-high
Cocuy sub-basin
Llanos Orientales
Moho
Palaeo-Central Cordillera
PALAEOGENE Palaeo-Magdalena Valley
Magdalena-Tablazo sub-basin
Tunja Axial Region
Cocuy sub-basin
Llanos Orientales
Moho
10 km (from this Chapter )
NEOGENE Magdalena Valley
Inverted Magdalena-Tablazo Magdalena-Tablazo Tunja Inverted Cocuy sub- Llanos sub-basin Axial Axial Region Region basin Orientales
105 km (from Colletta et al. 1990)
0
100 km
Moho
Approximate Approximate horizontal and and vertical scale
Figure 3.1: Cartoon showing the suggested tectonic evolution of the Eastern Cordillera since Mesozoic time. Results of this chapter indicate an incipient inversion of Mesozoic basins and generation of local topography within the study area (e.g. Gómez et al., 1999). This interpretation contrasts with the hypothesis of a single foreland basin during Palaeogene time (e.g. Coo per et al., 1995). 1995). Neogene and Cretaceous sections are the balanced and restored sections from Colletta et al. (199 0); however, sections s ections are not balanced at depth. Cretaceous shortening estimated from Cretaceous extension factors factors ( Chapter 2), Palaeogene shortening estimated from kinematic modelling (section 11 of this chapter) and Neogene shortening sh ortening from fromC olletta et al. (1990).
No subsidence studies or flexural models have been done for the Palaeogene of the study area. Several hypotheses about tectonic vertical movements are tested through 1D-thermal subsidence modelling and 2D-flexural modelling to explain the observed tectonic subsidence. An extensive data set of more than
73
Chapter 3
100 stratigraphic columns and wells from the EC, MV and LLA areas ( Fig. 3.2) f rom rom literature and well data from Ecopetrol was used. Because lithosphere stretching g enerated the the basin during the Mesozoic, the first hypothesis tested was whether some remaining thermal subsidence after Mesozoic stretching affected the Palaeogene evolution, which is a similar scenario to the Aquitanian Basin, that was quantitatively tested by Deségaulx et al. (1991). The second hypothesis tested was lithosphere flexure produced by a topographic load in the palaeo-Central Cordillera, as proposed in literature ( e.g. Cooper et al., 1995). The third hypothesis tested was flexural subsidence produced by the topography generated by some local uplifted blocks within the study area. This last hypothesis was considered by taking into account evidence of active deformation, slight inversion (Gomez et al., 1999) and surfaceuplift of hanging-wall blocks adjacent to Mesozoic extensional faults during the Palaeogene. Although the observed subsidence could not be explained by any individual hypothesis, a combination of all effects gave a satisfactory result. The observed local flexural component of subsidence is interpreted as having been produced by surface-uplift of local faulted blocks. To better understand relationships between vertical and horizontal horizontal movements, movements, horizontal horizontal compression compression movements movements responsible responsible for rockuplift of local faulted blocks were calculated through forward modelling of kinematics, structural deformation, sedimentation and erosion assuming an incipient inversion of the Mesozoic extensional basin during Palaeogene Palaeogene (Sassi et al., 1998). This model allowed an estimate to be made of the amount of basin inversion and shortening during Palaeogene.
2. TECTONIC SETTING 2.1. PLATE-TECTONIC INTERPRETATIONS 2.1.1. L atest atest Cr etaceo etaceous us and Terti ar y
During latest Cretaceous (since Campanian) and Tertiary all published plate-tectonic interpretations interpretations (Pindell and Dewey, 1982; Burke et al., 1984; Duncan and Hardgraves, 1984; Ross and Scotese, 1988; Jaillard et al.,1 990; Pindell and Barret, 1990; Pindell, 1993; Pindell and Erikson, 1993; Meschede and Frisch, 1998) propose that a convergent margin existed west of Colombia. During latest Cretaceous to Oligocene the Caribbean plate was moving eastward relative to South America, while the Farallon Plate was subducting west of Southern Colombia (Pindell and Erikson, 1993 and Pindell and Tabut, 1995). Late Cretaceous-Early Paleocene. There is agreement about the subduction of a south-western portion of a Caribbean or proto-Caribbean oceanic plate west of the Central Cordillera and obductionobductionaccretion of oceanic terranes (Calima Terrane according to the terminology of Toussaint and Restrepo, 1989; Toussaint, 1995a,b) to form the Western Cordillera (McCourt et al., 1984). Caribbean collision with north-western South America was diachronous, becoming younger northward and eastward: Cenomanian-Campanian in Ecuador and Campanian-Maastrichtian in Colombia (Pindell and Erikson, 1993 and Pindell and Tabut, 1995). According to Cooper et al., (1995) deformation was restricted to the Western and Central Cordilleras except for initial rock-uplift of the Santander Massif (Fabre, 1987). The amount of compressional deformation generated during accretion may have been limited by the oblique convergence of the Nazca and South American plates until 49 Ma (Pardo Casas and Molnar, 1987). According to Cooper et al. (1995) an early-Pre-Andean foreland basin was developed in the area of the EC, MV and LLA. Eocene. An increase in rate of Caribbean-South American plate convergence between 49 and 42 Ma (Daly, 1989) elevated the Central Cordillera and produced folding and thrusting in the Middle MV. These structures are truncated and unconformably overlain by upper Eocene clastics (Morales and
74
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes a s L u c a n S d e
) y r a d n u o b
1400
t e l a p
1300
i a n a r r e S
e r e ( F
Serrania de Perija
Maracaibo Basin Cucuta
s d e n A a d i e r M
Barinas Basin
Fig.3.15
F g i . b a 3 .1
0
100
200 km
Bucaramanga Arauca
1200
Medellin
a e r l l i d o r C
1
Llanos Orientales Basin (LLA)
2 Tunja
1100
Yopal
Manizales
3 1000
900
800
Bogota
re Ibague t u u S l ra e m o R
F i g .3 b .1 a 4 F g i . 3 .9 Villavicencio
Fig.3.8
l r a t e n C
800
5
Guyana Shield
e a d n e a r i a n c a a r r M e a S L
Neiva
900
1000
8
San Jose del Guaviare
1100
1200
1300
1400
1 2
1500
Figure 3.2: Location of stratigraphic columns, wells used, and loca tion of lithosphere flexure 2Dmod el sections. Numbers along sections refer to labelling of stratigraphic transver sal sections (Figs. 2.7, 2.8 a nd 2.9), and also to labeling of lithosphere flexure 2D models (Table 2.1; Figs. 3.26 to 3.33). The Romeral suture i s the westernmost boundary of continental crust. It was used as a free plate boundary f or the 2D flexural models
the Colombian Petroleum Industry, 1956). The Eocene unconformity is regional in eastern Colombia. According to Cooper et al. (1995) a late-Pre-Andean foreland basin developed in the area of the EC, MV and LLA from Middle Eocene to Early Miocene time. Latest Eocene and early Oligocene. The rate of plate convergence was reduced, from about 2 cm/yr to 1 cm yr. Convergence mainly was driven by South America moving westward toward a nearly stationary Caribbean Plate, referred to a mantle reference frame. Sedimentation reassumed in eastern Colombia possibly by relaxation of the greater compressive stresses in early and middle Eocene times (Pindell et al., 1997).
75
Chapter 3 Late Oligocene-Early Miocene. The
Farallon Plate was broken into the Cocos and Nazca Plates during Oligocene at about 25 Ma ago (Wortel and Cloetingh, 1981; Duncan and Hardgraves, 1984). According to Cooper et al.( 1995), changes in plate-tectonic motions (Pilger, 1984; Avraham and Nur, 1987) did not cause any deformation in the EC or the LLA. Deformation of this age has been described from the Cauca Valley (Alfonso et al., 1989) and the MV where the reactivation of the middle Eocene structures created an upper Oligocene unconformity (Schamel, 1991). Middle Miocene. Collision and accretion of the Cuna Terrane (Serranía de Baudó in NW Colombia) with the northwestern margin of South America occurred during the Middle Miocene (Duque-Caro, 1990). According to Cooper et al. (1995) the collision may have contributed to loading and may have initiated deformation in the EC. However Kerr et al. (1997), based on the age of the Dabeiba subduction-related magmatic arc, suggested that obduction and accretion of the volcanic rocks of the Serranía de Baudó (Cuna Terrane) probably occurred during late Eocene. A similar westward shift of the subduction zone during Eocene has been proposed by Feininger (1980, 1986) in Ecuador.
3. STRATIGRAPHY In this section I summarise the Palaeogene stratigraphy ( Figures 2.7 to 2.11). Because the Palaeogene tectonic history of the EC is not well constrained, I also pres ent in this section stratigraphical ( e.g. unconformities revealing local erosion), sedimentological ( e.g. palaeocurrent data suggesting location of detrital provenance areas) and petrographical (mineralogical detrital sandstone composition suggesting detrital provenance areas) evidence, useful to constrain the Palaeogene tectonic history.
3.1. LATE MAASTRICHTIAN-EARLY PALEOCENE According to Cooper et al. (1995) the final accretion of the Western Cordillera ( Figs. 1.1 and 1.2) resulted in a change from marine to the non-marine deposition in the pre-Andean foreland basin (Figs. 2.7 to 2.11 a nd Figs. 3.3 and 3.4). During late Maastrichtian-early Paleocene, paralic coastal plain and alluvial plain claystone s containing coal beds of the Guaduas Fm were deposited (Laverde, 1979; Fabre, 1985a; Sarmiento, 1992, 1993). The lower limit of the Guaduas Fm is defined by a generalized marine flooding surface (Sarmiento, 1993), and upwards a general regression and change from coastal marine environments to alluvial plains is recorded. To the northward facies are more marine (Catatumbo Fm, Fabre, 1985a; Sarmiento, 1993). Sarmiento (1993) recognized four smaller sequences within the Guaduas Fm. Each sequence starts with a marine flooding surface an d records a change from coastal marine environment to alluvial plains. Based on thickness change s (Fig. 3.3) Sarmiento (1993) suggested that faults controlled the sedimentation, with the most subsi ding blocks located in the western flank of the Cordillera. The Fm is limited on top by an uncon formity (Sar miento, 1993), which is inter preted as an erosional truncation increasing in effect eastward ( Figs. 2.7 to 2.11). In the LLA Foothills (Figs. 2.1), upper Maastrichtian-lower Paleocene rocks are not present (Cooper et al., 1995). In the U pper MV (Fig. 2.1) mudstones and sandstones (Seca Fm, Guaduala Fm) were deposited by meandering streams in a delta plain environment during the marine regression (Gómez and Pedraza, 1994). The basin was completely full of sediment by the end of early Paleocene (Sarmiento, 1993). In the EC the Guaduas Fm is conformable over the Guadalupe Gp or equivalent units. However, Guillande (1988) reported that locally in the Upper MV the Guaduas Fm. rests unconformably on rocks as old as Jurassic to Upper Cretaceous, suggesting a Late Cretaceous deformation event in this area.
76
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera ,Colombian Andes
1400
Serrania de Perija
s a c u L n a S e d i a n r a r e S
1300
Cucuta
>550 .
a a r e l i d o r C
a p
Medellin
a a c o y665. B
200
Tunja
1100
850. 810. 559.
309.
0
>800 .
t l u a f >250 o i . e v l a l t u p a a f z o r e a a e l u a p S
t
0
>100.
0.
100
0
760. 657.
1000 Ibague
>115 . 131.
900
198.
>40 >65 . . >50 >107 . . >45 . 133. >144 .
458.
0. 0.
r e S
320.
390. 360.
530. >242.
575.
660. 450.
817. 540.
o a e l p a o e m a m y s t 0. r á i c l s u a f a u 0. G
0.
Llanos Orientales Basin (LLA)
0.
0.
0. 0.
0.
140.
Yopal
0. 0.
1150. 130.
>258 .
90.
Bogota
0.
130.
0.
510.
0.
478.
0.
0.
400.
0.
0. 0.
Villavicencio 0.
>61 . 70.
0.
0. 0.
0.
800
0.
0. 0.
0.
0.
0.
MAASTRICHTIAN EARLY PALEOCENE
a e n e d r a a i n c a a M r r e a S L
30.
800
0.
335.
Neiva
900
1000
0.
0.
0. -
l r a t e n C
Arauca
374.
1400. >345 .565.
Manizales
Barinas Basin
942.
0.
1200
i d a r e M
s d e n A
Bucaramanga
100
a l
Maracaibo Basin
0
100
1100
THICKNESS (metres)
200 km
1200
1300
1400
1500
Figure 3.3: Maastrichtian-early Paleocene restored thickness (meters) without palinspastic restoration. Large bold numbers indicate palaeotopography in meters estimated from 2D flexural modelling. Thick lines represent palaeofaults believed to have been active during early Paleocene time.
The cycle ended with a local unconformity due to slight exhumation of some areas, probably related to the initial deformation of the sedimentary pile. Subtle exhumation and erosion resulted from the initial inversion of the normal faults at the borders of the basin during early Paleocene (Fajardo-Peña, 1998). Evidence of this deformation is the unconformity at the top of the Guaduas Fm. Detrital source areas. During Maastrichtian time rocks of the Central Cordillera ( Fig. 1.1) began to uplift and supply detrital sediments in some places along the western margin o f the basin (quartzite, chert pebbles in the Cimarrona Fm, Gómez and Pedraza, 1994). Also the mineralogical composition of sandstone in the Upper MV (metamorphic fragments in the sandstones of Monserrate Fm, Waddell, 1982, and in western Venezuela, Van Andel, 1958) indicate exhumation of the Central Cordillera. Although most of the Maastrichtian-early Paleocene Guaduas Fm sandstones are quartzrich, in its upper part, sandstones are texturally and compositionally immature, having also been derived from erosion of the Central Cordillera (Sarmiento, 1993).
77
Chapter 3
1400
ra e i ll rd o C l a i tr n e Medellin C o le a P
1300
a t
1200
1100 Manizales
1000
900
m t e s y s l t u f a o e la a p a m s u h C
a i n o n d B as H b S u Ibague
t o a d i s n r a a r B i - y G u b l e l S a
Neiva
800
V a n e l a d g a M r e p p U
h ig H r a c hi a C y e
900
Maracaibo Basin Cucuta
i d a r e M
s d e n A
I
n v
a l t i n f a u l a S e o a a L l t p a u l f a e S o e n o i z a s l a o l a B b o s p a p a a b T m z a d S u c a r e o g i o n r t e n a o y a a S e l s u v a l e B h i j a R e g S t I n g d n o o a T u M s F o a n e m L l y s t t a Tunja o g i n l s Yopal u a s B o a B f b d e e o S u a a l y n c u b a p a o C o S a v m I n r a Bogota i c a a G u
i f s a s M e m a t u e Q
Arauca
Llanos Orientales Basin (LLA)
PALEOCENE PALEOGEOGRAPHY
Positive relief. Alluvial fan and f luvial Villavicencio
0
1000
Barinas Basin
Bucaramanga
l l e V a l d i d n a l t M l e a rv a u d f a g e o M l a
a e n e d r a i a n c a a r M r e a S L
v a i n i e a s N B b S u
800
s a c u L n a S e d i a n r a r e S
S a n t a n d e r M a s s i f
100
200 km
San Jose del Guaviare
1100
1200
1300
Coastal plain predominantly sandstones Coastal plain predominantly mudstones Littoral to inner shelf sandstones Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltstones Outer shelf shales of carbonates Turbiditic sandstones
1400
1500
Figure 3.4: Paleocene palaeogeography without palinspastic restoration (modified from Geotec, 1992; Cooper et al., 1995 and Villamil, 1999).
3.2. LATE PALEOCENE During late Paleocene time i n the LLA region and EC (Fig. 1.1) sedimentation started after a period of erosion ( Figs. 2.7 t o 2.11, Sarmiento, 1993). Abrupt changes of thickness of the Guaduas Fm (e.g. across the Soapaga Fault with a thinner section in the eastern side, Fig. 3.5) suggest some faults moved during the time represented by the unconformity (Céspedes and Peña, 1995). In the LLA foothills (Figs. 2.1 and 3.4, near Paz de Ariporo) locally the unconformity has been described as angular (Vanegas and Arango, 1994). In the Middle M V (Figs. 2.1 and 3.4) an onlap relation of Paleocene deposits over older rocks can be recognized on seismic lines (Ecopetrol-ICP, 1996) and might be the expression of the unconformity. Upper Paleocene deposits form a sequence bounded by unconformities. The upper boundary is an unconformity that represents the early–middle? Eocene recognized in Eastern Colombia (Figs. 2.7 t o 2.11). In the LLA ( Fig. 1.1) this sequence extends farther east, possibly due to transgression and early flexur al loading of the Centr al and Western Cordilleras (Cooper et al., 1995). The lower basal transgressive system tract is represented in the LLA by mature, sandstone-rich, estuarine deposits (Barco Fm, Cooper et al., 1995).
78
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes Serrania de Perija
s a c u L n a S e d i a n ra r e S
1400
Maracaibo Basin Cucuta
i d a r e M
s d e n A
Barinas Basin
500
1300
1100
Bucaramanga
100
314
ra e i ll rd o C
1200
100
>223
0
Arauca
964
100
0
458
1007
0
Medellin
951
7 2 8
100
24
859
0
100
1100
560
100
0
200
1170
708
560
610
1385
105
50
155
Ibague
200
Bogota
0
76
270
54
Yopal
200
3617
280 530
180
1500
56? 109 96 29 95 72 16
12 0
0
0
0
0 0
47
0
0
24
0
68
0
0
0
31 0
Villavicencio
>58 75
0 0
60?
0
0
00
0
PALEOCENE 0
THICKNESS
415
Neiva
800 0
800
900
a e n d e r a a i n c a a M r r e a S L
1000
0
0
0 77
a l n t r e C
0
0
145 >173
900
Llanos Orientales Basin (LLA)
0
>443
Tunja 708
950
100
0
660
0
0
0
904 666 525 >387
480
70?
Manizales
1000
0
834
0
(meters) 100
0 1100
1200
200 km
1300
1400
1500
Figure 3.5: Paleocene restored thickness (meters) without palinspastic restoration. Large bold numbers indicate palaeotopography in meters estimated from 2D flexural modelling. Thick lines represent palaeo-faults believed to have been active during Paleocene time.
In the EC this lower part is represented by braided fluvial sandstones (Thanetian age, Sarmiento in Jaramillo et al., 1993; Arenisca del Cacho and Socha Inferior Fms, Céspedes and Peña, 1995), and in the Middle MV by a sandstone stratigraphical interval (basal part of the Lisama Fm considered by Bueno, 1971 in Mora et al., 1996) to be lateral equivalent of the Barco Fm in the Catatumbo area). Marine influence is strong in the LLA foothills (Cooper et al., 1995) while in the EC a deltaic coastal plain with marine ichnofauna represents a maximum flooding surface (lower part of Socha Superior Fm, Céspedes and Peña, 1995). The transgressive system tract is represented in the LLA by the transition from estuarine deposits and the highstand system tract is represented by an upward transition to coastal (lower Los Cuervos Fm) and alluvial plain mud-rich deposits (Cooper et al., 1995). Sandstone deposition ended when a relative tectono-eustatic level highstand was established (Cooper et al., 1995). Coarse clastics appear to have bypassed the LLA foothills and EC, where a regressive mud
79
Chapter 3
dominated coastal (Cooper et al., 1995) to alluvial plain with high sinuosity streams was established (Bogota Fm, Hoorn, 1988; upper part Socha Superior Fm, Céspedes and Peña, 1995). In the Middle MV shallow marine to transitional variegated mudstone, minor sandstone and thin beds of coal in coarsening upward successions are interpreted as produced by decrease of the accumulation space/sediment supply ratio (Ecopetrol-ICP, 1996). Northward in the Sierra Nevada del Cocuy the environment was more marine (Barco Fm with littoral environment, Fabre, 1986). For the LLA area, Fajardo et al. (1993) have proposed that this sequence bounded by unconformities represents incised valley fills deposited during a base-level rise trend formed by four smaller base-level cycles. The lower part consists of fluvial sandstones (Barco Fm) overlain by transitional deltaic(?) varying to fluvial de posits eastward (Los Cuervos Fm). Thickness variations ( Fig. 3.5) within this sequence are interpreted by Cooper et al. (1995) to result from extension on the Cusiana-Támara fault system in the LLA foothills. In the EC also changes in thickness have been recognized (Céspedes and Peña, 1995). Detrital source areas. Upper Paleocene sandstones (Arenisca del Morro, Barco, Socha Inferior and Arenisca del Cacho Fms) are quartzarenites (Jaramillo et al., 1993; Cooper et al., 1995,), litharenites, sublitharenites (Jaramillo et al., 1993, Barrientos and Torres, 1994) and subarkoses (Jaramillo et al., 1993). Sandstones contain quartz similar to that of Upper Cretaceous Guadalupe Gp (Barrientos and Torres, 1994), chert fragments (Notestein et al., 1944; Aalto, 1971; Vásquez, 1983; Hathon and Espejo, 1997; some of them derived from La Luna Fm, Fabre, 1986) and low-grade metamorphic clasts (Hathon and Espejo, 1997). Palaeocurent data from these sandstones are: (1) Northward to northwestward (Jaramillo et al., 1993, Barrientos and Torres, 1994) in the LLA foothills (Fig. 3.4); (2) Southeastward in the northwestern axial region of the EC (Fig. 3.4, Jaramillo et al., 1993); (3) Northeastward in the Sabana de Bogotá area (Fig. 3.4, Jaramillo et al., 1993); (4) Northward in the Tunja Paz de Río area (Fig. 3.4, Céspedes and Peña, 1995); (5) Toward the north in the TunjaPaz de Río area east of the Soapaga Fault (Céspedes and Peña, 1995), (6) Toward northwest in the Tunja-Paz de Río area west of Soapaga Fault (Céspedes and Peña, 1995). Based on detrital composition and palaeocurrents data the following detrital source areas have been proposed: (1) Santander-Floresta Massiff ( Fig. 3.4, Fa bre, 1986; Jaramillo et al., 1993); (2) Quetame Massif (Fig. 3.4. Hoorn, 1988; Jaramillo et al., 1993; Barrientos and Torres, 1994); (3) The Guyan a Shield (Fig. 3.4, Rincón 1982 in Barrientos and Torres, 1994; Jaramillo et al., 1993, Cooper et al., 1995); (4) The Centr al Cordillera: proposed only for the sandstones of the Middle Magdalena area (Fig. 3.4, Lisama Fm, Hathon and Espejo, 1997). Detrital composition and palaeocurrent data suggest that local erosion of the Upper Cretaceous sedimentary rocks of the EC supplied quartz and fragments of chert and mudstones and generated local exhumated areas that controlled deposition (Fabre, 1986; Jaramillo et al., 1993; Barrientos and Torres, 1994; Hathon and Espejo, 1997). Evidence presented here indicates that some areas of the EC started to exhumate and to supply sediment indicating that the early pre-Andean foreland basin proposed by Cooper et al. (1995) was segmented by local exumated areas.
3.3. PALAEOGENE (EOCENE TO EARLY MIOCENE) According to Cooper et al., (1995) the Eocene to Early Miocene sedimentary record was deposited in a late pre-Andean Foreland basin. Because most Tertiary sediments of the MV and some of the EC and LLA are dominantly continental and generally poorly fossiliferous, there are no reliable ages for Palaeogene rocks in the whole MV (Jordan and Gomez, 1996) and correlation with the EC or LLA is not well established. Thus the summary will proceed separately for different areas.
80
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes 3.3.1. LLanos and Eastern Cordillera L ate Eocene
These rocks are bounded at the bottom by the regional early to middle? Eocene unconformity (Figs. 3.6 to 3.9). The unconformity in some places has been described as a paraconformity, but in others places, at least locally, as an angular unconformity ( e.g. locally in the LLA foothills (Aguaclara area) the lower contact of the Mirador Fm is an angular unconformity over the Cuervos Fm., Barrientos and Torres, 1994). Fabre (1986) has reported a local angular unconformity at the bottom of the Middle (?) to late Eocene Picacho Fm in the Paz de Río area. B Serrania u c a de Perija r a m a n g a (?) F
s a c u L
1400
F l a r e m o R
1300
n a S e d ia n ra r e S
(?)
r a l e l i d o r C
Arauca
F e z r a S u
Medellin (?)
Tunja
1100 Manizales
F s r a b m a C
(?) Bogota
1000
m t e Yopal y s s l t u f a o m a r a c i a G u
Ibague
l F e ra om R 900
S
Llanos Orientales Basin (LLA)
F n a i l S a L a
1
Barinas Basin
C Cucuta h i t a g a F
Bucaramanga
e
1200
Maracaibo Basin
Villavicencio
l r a t e n C
0
100
200 km
EARLY EOCENE UNCONFORMITY Beheaded accretionary wedge Highly variable angular unconformity
(?)
Truncated homocline / broad low relief folds
(?) Neiva
800
a F m s h u C
800
F i r a m l t a A
900
Uplifted blocks with no deformation or very modest deformation
e a d n e a r i c n a a a r r M e a S L
1000
(?) 1100
1200
Paraconformity Evidence 1300 eroded
1400
1500
Figure 3.6: Nature of the early Eocene unconformity. Without palinspastic restoration (modified from George et al., 1997).
In the LLA area ( Fig. 1.1) initial deposition consisted of marine-influenced, s andstone-rich, fluvial and estuarine valley fill deposits, and continued in muddier coastal plain sediment s (Figs. 2.7 to 2.11, 3.10, Pulhamn, 1994; Cooper et al., 1995; Fajardo, 1995). Coarse and often p ebbly, fluvial
81
Chapter 3
and alluvial fan sandstones are the dominant sediments deposited over a wide area of the LLA area (Mirador Fm). Continued transgression eventually submerged the alluvial plain and established a shallow marine shelf across the Cusiana area (Cooper et al., 1995). The upper part comprises heavily bioturbated estuarine parasequences punctuated by sandstone-rich, estuarine valley fill deposits (Cooper et al., 1995). In the Cusiana oil field major flooding ended sand deposition (Fajardo et al., 1993; Fajardo, 1995; Cooper et al., 1995). In the axial region of the EC a braided fluvial fining upward succession with conglomerate at the bottom followed by coarse to medium grained sandstone (Regadera Fm, Julivert, 1970; Picacho Fm, Giraldo et al., 1993) recorded a base-level rise. A maximum flooding surface possibly is represented by marine iron-rich oolític facies in the lower part of the Concentración Fm. (Céspedes and Peña, 1995; late Eocene according to Cazier et al., 1996). This sequence is only present in the western side of the Soapaga Fault. Significant lateral changes of thickness may be the result of fault control (Cooper et al., 1995). 4 Km
NW 700
600
500
400
SE 300
200
100
0.0 0.5 1.0 1.5
2.0 2.5
K
3.0 3.5
) 4.0 c e S ( T 4.5 W T
Figure 3.7: Seismic line BR8609-SC through the Cáchira High in the Middle Magdalena Valley. Note the lowermiddle? Eocene angular unconformity truncating earlier structures (from Mantilla, 2000). Location of this line is indicated in figure 3.2.
In the LLA area sandstones are extremely mature quartz arenites, although locally fine-grained litharenites occur (Fajardo et al., 1993; Fajardo, 1995; Cooper et al., 1995). In the EC (Tunja-Sogamoso area) conglomerates contain clasts of black chert, some with Cretaceous foraminifera, sedimentary and low grade metamorphic rocks, evidencing erosion of some areas of the EC( Figs. 3.10 a nd 3.11), as also suggested by palaeocurrent data (toward SW, Céspedes and Peña, 1995). Northward in the C ocuy area the unit contains conglomerate and lithic to sublithic sandstones with fragments of Cretaceous chert, mudsto nes and polycr ystalline quartz (Fabre, 1986). Fabre (1986) proposed the uplift of the Santander Massif (Figs. 3.10 and 3.16) as detrital source. Detrital source areas.
82
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes
According to this author, the relative decrease in chert content compared to late Paleocene units suggests that the erosion reached deeper levels than the Late Cretaceous chert eroded during late Paleocene. Based on detrital mineralogical composi tion of sandstones, in the Tunja-Sogamoso area Mejía and Giraldo (1993) concluded that the detrit al source area was located in the Floresta Massif and probably in the Santander Massif ( Figs. 3.10 and 3.11). Based on sandstone composition and palaeocurrent data (toward NNE) Barrientos and Torres (1994) proposed the Guyana Shield and the Upper Cretaceous sandstones of the EC as detrital sources for the LLA foothills (Aguaclara area). Evidence presented here indicates that some areas of the EC started to became exhumated and to supply sediment. Once again local exhumated areas segmented the late pre-Andean foreland basin proposed by Cooper et al.( 1995).
Figure 3.8: Outcrop view of the lower-middle? Eocene angular unconformity truncating Cretaceous beds. On top the Palaeogene Chicoral Fm (part of the Gualanday Gp.) Drawing from a photograph looking at the South, taken in the Loma de Los Caballos, U pper Magdalena Valley (from Amézquita and Montes, 1994). Location of this outcrop is indicated in figure 3.2. Oli gocene to Earl y M iocene?
In the LLA areaa nd foothills four cycles of marine influenced, lower coastal plain deposition accumulated ( Figs, 2.7 to 2.11 and 3.12, Carbonera Fm). These sequences recorded eastern migration of foreland basin subsidence, which culmi nated with the onset of EC deformation (Cooper et al., 1995). Eastward these sequences are sandier and more continental approaching the Guyana Shield. Sequences thicken westward ( Fig. 3.13). Cooper et al. (1995) suggested that continued episodic normal displacement on the Cusiana fault system took place during this time. Each sequence consists of a thin, forced regression systems tract, a sand prone transgressive systems tract, a maximum flooding surface and a mud-dominated highstand system tract (Cooper et al., 1995). In the axial region of the EC correlative deposits are yellow to grey claystone with thin beds of medium grained sandstone (Concentración Fm, Reyes and Reyes, 1976, Oligocene-early Miocene according to Germerard, 1968; Usme Fm, Julivert, 1970). Northward in the Cocuy area, Fabre (1986) described the lower part of the Concentración Fm, as containing dark grey to black locally carbonaceous mudstone with thin beds of
83
Chapter 3
sandstone, one of them with iron oolites, which he interpreted as having been deposited in a littoral to shallow marine environment. Cooper et al. (1995) stated that sediments prograded westward and onlap eastward. Julivert (1970) described local unconformities in anticlines but continuous sedimentation in the Usme Syncline in the Regadera and Usme Fms of the Sabana de Bogotá area and interpreted this as the result of simultaneous folding and sedimentation. ESE
WN W 6
1
2
3
4
5
6
Figure 3.9: Section constructed from field data through the Carmen de Apicalá Syncline in the Upper Magdalena Valley. 1. Cretaceous Villeta Fm. 2. Upper Cretaceous Guadalupe Gp. 3. Palaeogene middle part of Gualanday Gp. 4. Palaeogene Upper part of Gualanday Gp. 5. Palaeogene? Barzaloza Fm. 6. Neogene Honda Gp. Note unconformities and lateral changes of thickness (from Guillande, 1988). Location of this section is indicated in figure 3.2.
According to Cooper et al. (1995) the main source of sediments was the Guyana Shield ( Fig. 3.12). However, Cardona and Gutierrez (1995) studied the mineralogical detrital composition of the sandstones (sublitharenites to muddy sublitharenites and quartzarenites at the bottom) of the Carbonera Fm in the LLA foothills (NW of Yopal). They found metamorphic rock fragments (phylite quartzite), plagioclase, microcline, chert, polycrystalline quartz, zircon and tourmaline, feldspar and locally igneous rock fragments. Metamorphic fragments generally are phylite indicating a relative near detrital source area. Well preserved, angular up to medium sand size grains also indicate a near detrital source area. Palaeocurrent data dominantly are eastward (SSE and ENE). They conclude that at least two detrital source areas existed which were located at the west, one of them near the deposit. In the detrital source areas both metamorphic-igneous and sedimentar y rocks were eroded. The nearby area could be located at the Floresta, Santander or Quetame Massi fs (Fig. 3.12) and probably includes some erosion from Late Cretaceous sedimentary rocks from the EC (Guadalupe Gp), as suggested by the presence of chert and polycrystalline quartz fragments. The far detrital source area could be the Central Cordille ra (Fig. 3.12, Cardona and Gutierrez, 1995). Moreno and Velázquez (1993) described an upward increa se in sand grain size and interpreted it as reflecting a slight exhumation of the detrital source area. The presence of Cretaceous chert clastics with foraminifera (Fajardo-Peña, 1998) corroborates that some parts of the EC started to exhumate and supply sediment, thus the late pre-Andean foreland basin propose d by Cooper et al., (1995) was segmented by local exhumated areas. Seismic evidenc e (Fig. 3.14) suggests that contractional inversion of Mesozoic faults occurred during Palaeogene. Detrital source area s.
3.3.2. M agdalena V alley (M V)
Tertiary sediments of the MV ( Fig. 1.1) ar e dominantly continental and poorly fossiliferous (Figs. 2.7 t o 2.11 a nd Figs. 3.10 to 3.13). There are no reliable ages for Palaeogene rocks in the whole MV.
84
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes EOCENE PALEOGEOGRAPHY 1400
Positive relief. Alluvial fan and fl uvial
1300
1200
1100
1000
900
800
Coastal plain predominantly sandstones Coastal plain predominantly mudstones Littoral to inner shelf sandstones Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltstones Outer shelf shales of carbonates
s a c u L n a S e d ia n y r ra l e a e V S a n l e a d a g M l e d i d M
Serrania de Perija S Cucuta a n t F I n a n l o v d r e S e r s a M ta n t a a s H n d s ig e i f h r
Maracaibo Basin Barinas Basin
Arauca
Bucaramanga
n a l e a d a g M n i o s z a B a l Turbiditic sandstones b b T a u d S t o ra m l t e u o s il e s te r a e f d v g a m n s y or I n a e o Tunja m o C lt l S o eg i y s t e p a t ra l fa u n j a R l t s e z eo a r u Yopal e n u f a Tu C S la e o Manizales d p a eo a a l a e l n a a la a t a a p u t i n b g o n d a s S a P o f a u im o o B t H b b a m o B i á r a l a e u c i S a p Bogota G u c a Ibague a t n o y o B i y r d a s Villavicencio l e i r a b a G u V b a S n l e d a g a M e r o p p e U
m l a e a t Neiva p y s a s m s n a i t u l s u i v a h a e C f N b
e a d n e r a a i n c a a r r M e a S L
b u S
800
900
1000
San Jose del Guaviare
1100
1200
l s i h t o o F s n o a l L
Llanos Orientales Basin (LLA)
Guyana -
0
Shield
100
1300
200 Km
1400
1500
Figure 3.10: Eocene palaeogeography without palinspastic restoration (modified from Geotec, 1992; Cooper et al., 1995 and Villamil, 1999).
These dominant continental rocks are characterized by important lateral changes of facies and thickness. In the Middle MV ( Fig. 2.1) from bottom to top the units are the Chorros Gp (La Pa z, Esmeraldas Fms) and the Chuspas Gp (Mugrosa and Colorado Fms). In the Upper MV ( Fig. 2.1) the major units from bottom to top are the Gualanday Gp and the Barzaloza Fm. In the Ho nda sub-basin (Fig. 2.1), the Hoyón and San Juan de Rioseco Fms have a similar relative stratigraphic position as the Gualanday Gp. However, in the Honda sub-basin, Porta (1965, 1966) showed that the conglomeratic Hoyón Fm pinches out and disappears totally toward the south and east of the Guaduas syncline and thus, it is not a northward lateral continuation of the Gualanday Gp of the Upper MV. In the Honda sub-basin the Santa Ter esa Fm, which lithologically overlies these units, is chronostratigraphically similar to the uppermost part of the Chuspas Gp of the Middle Madgalena Valley (La Cira fossiliferous shale horizon, Nuttal, 1990, Figs. 2.7 t o 2.11 an d Figs. 3.10 to 3.13).
85
Chapter 3
1400
s a c u L n a S e d ia n r ra e 67 S
EOCENE
THICKNESS (meters) 1300 0
r a e l l i d r o C
1200
s d e n A a d r i e M
Cucuta
Barinas Basin
200 935
Bucaramanga
t l u f a o a e l a p z r e a S u
39
Medellin
68 78
100
220
80
284?
45 284
53
50 160
40?
Llanos Orientales Basin (LLA) 0
32
Tunja 489
0
Arauca
200
l t f a u o l a e 380 a p a 250 c y a 186 80 B o
1150
0
1100
Maracaibo Basin
Serrania de Perija
Yopal
27
75
43
79
345
33 230 946 680
Manizales
1000
300
824 660
40? 86?99? 110 48 136 62 340 61 71 160 105 76 132 53 51 100 38 43 27
Villavicencio 490
173
48
38 28 23
77
27 35?
38
43
61
73
0
36 49
9
Guyana
0
a e n e d r a c a a i n a M r r e a S L
800
9 00
1000
0
28
0?
746
80 0
0
0
0?
Neiva
0
7 24
45 1150
2
20
92
294
l r a t n e C
76
29 28
Bogota 64
90 0
0 45
0
San Jose del Guaviare
1100
1200
0
Shield
100
1300
200 Km
1400
15 00
Figure 3.11: Eocene restored thickness (meters) without palinspastic restoration. Thick lines represent palaeo-faults believed to have been active during Eocene time. M iddle M agdalena V alley
The Tertiary record ( Figs. 2.7 to 2.11) is composed of cross-bedded sandstone, locally conglomeratic (fluvial channel), variegated to red mottled siltstone and mudstone (alluvial plain, lakes, and palaeosoils), dominantly of fluvial to lacustrine origin (Figs. 3.10 t o 3.13). Three major units with different sand /mudstone ratios are recognized, each one limited at the bottom by an unconformity. Each unit was deposited during a continuous increase of the accommodation space/sediment supply ratio that ended with lacustrine sedimentation (Ecopetrol-ICP, 1996). Lacustrine facies are fossiliferous green to black shales with fresh water to brackish mollusks and locally glauconite ( e.g. top Mugrosa Fm, Morales et al., 1956). Three of these fosiliferous shales lie at the uppermost part of each unit: (1) The Chorro Group (sand-rich La Paz Fm, and Esmeraldas Fms); (2) The Mugrosa Fm (mud-rich); and
86
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes Serrania Maracaibo de Perija Basin S a n t a Cucuta n d e r M a s s if
OLIGOCENE 1400
EARLY MIOCENE PALEOGEOGRAPHY
s a c u L n a S e d i a n a r r e S
Positive relief.
1300
1200
Alluvial fan and f luvial Coastal plain predominantly sandstones Coastal plain predominantly mudstones Littoral to inner shelf sandstones Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltstones Outer shelf shales of carbonates Turbiditic sandstones
1100 Manizales
Bucaramanga
y l e a V a n l e d a g a M l e d i d M
l r a t n e r a C a i n l e i o d n d a s r o a e l o H b b a C P u
S
o a z l b a T a l e n i n d a a s e d g t B r a e M b v u I n S
F l o r e s t a H i g h
Ibague
900
Neiva
t d o s i n r y a l e i r a b a G u b V S na
Arauca
Llanos Orientales Basin (LLA)
Tunja Yopal
d e n a t a f a b i S a B o g o a s s
M e m t a u e Q Villavicencio
i l l s t h o F o s o a n L l
le g d a a M er p p U
Guyana Shield a e n e d r a a i n c a a M r r e a S L
a i n i v a s e b N b u S 800
I n v S a n ta n d e r
u y n i o c s C a B v b I n S u
Bogota
1000
800
e s d A n a r i d e M
900
1000
San Jose del Guaviare
1100
1200
-
0
100
1300
200 Km
1400
1500
Figure 3.12: Oligocene-Early Miocene palaeogeography without palinspastic restoration (modified from Geotec, 1992; Cooper et al., 1995 and Villamil, 1999).
(3) the Colorado Fm (mud-rich, deposited in a mandering fluvial to lacustrine environment, EcopetrolICP, 1996). Sand-rich units were deposited predominantly by braided fluvial systems and mud-rich intervals predominantly in meandering fluvial systems (Jordan and Gomez, 1996). Mud-rich facies were developed especially during maximal base-level. Important lateral changes of thickness or texture occur (Morales et al., 1956). Maximum thickness occurs near the EC foothills (Morales et al., 1956) and decrease considerably or reduce to zer o on palaeo-highs tha t were episodically active during sedimentation ( e.g. Chucurí-Casabe, Cáchira , Figs. 3.7, 3.11 and 3.13) separating sub-basins ( e.g. Cantagallo-Yariguí; Cross et al., 1996a,b). Several smaller sequences have been recognized by Ecopetrol et al. (1994), Cross et al. (1996a,b) and Ecopetrol-ICP (1997). Ecopetrol et al. (1994) recognized 6 sequences (one approximately equivalent to La Paz and Esmeraldas Fms, two approximately equivalent to Mugrosa Fm, and three approximately equivalent to Colorado Fm) which are richer in sand westward. According to Ecopetrol et al. (1994), 5 sequences display similar truncation and onlap patter ns (Fig. 3.15). Each sequence thins westward.
87
Chapter 3 Serrania de Perija
OLIGOCENE 1400
EARLY
Cucuta s a c u L n a S e d 100 i a n a r r e S
MIOCENE THICKNESS 1300
Maracaibo Basin
(meters)
e s d A n a r i d e M 0
100
950 1117
Barinas Basin
200 Km
Bucaramanga
2130
1034 952
>689
100
>64
464
Arauca
270 358
1350
285
140 315
Medellin
1200
0
r a i le d o r 100 C
100
410
1475
315
637
100
1500
Tunja
478 600
0
768
50 1050
Manizales
0
100
80
829
1000
Ibague
780
1040
650
250
Bogota
250
1010
582
1079
2615
710 671 698 701
1079
887
912
800
800
387
381 415 381 534 430 427 325
304 336
334
305
256
247 232
309
55
561 271
726
238
426
748
377 374 378
797
l n t ra e C
358 533
512
564 850
200
407
1026
Villavicencio
49
900
309
597
0 1 0 0
382
380
451
520
1621 425
Llanos Orientales Basin (LLA)
719
Yopal
1478 995 >301 991 665 1051 671
650
721
299
460
150
441
1100
405
1200
268 807
a e n d e r a i a n c a a r M r e a S L
Neiva
900
1000
Guyana Shield
227
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 3.13: Oligocene-Early Miocene restored thickness (meters) without palinspastic restoration. Large bold numbers indicate palaeotopography in meters estimated from 2D flexural modelling. Thick lines represent palaeofaults believed to be active during Oligocene-Early Miocene time.
The basal beds of each sequence onlap the underlying lower sequence boundary from east to west while the tops of each sequence are truncated by the overlying upper sequence boundary. These relations are the effect of regional tilting and accommodation with sediment having greater preservation potential away from the Central Cordillera. All sequences onlap the western flank of the basin in the southern Middle MV, they are fault-bounded against the Cantagallo Fault, and are exhumated and truncated in the northern part of the basin (Ecopetrol et al., 1994). Cross et al., (1996a,b) and Ecopetrol-ICP (1996) also recognized 6 stratigraphic intervals delimited by time correlation surfaces defined through recognition of base-level or accommodation space/sediment supply ratio cycles. Similar to the Ecopetrol et al.’s sequences, one interval is approximately equivalent to La Paz and Esmeraldas Fms, two approximately equivalent to the Mugrosa Fm, and three approximately equivalent to the Colorado Fm. The regional Eocene unconformity is the base of Palaeogen e sedimentar y record ( Figs. 2.7 to 2.11 and 3.6). In the Middle MV it is a dramatic angular unconformit y (Fig. 3.7). Paleocene-Cretaceous and older Fms below it are strongly folded and faulted. This unc onformity post-dates
88
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes
most but not all exhumation and folding (Cross et al., 1996a,b). According to Cross et al. (1996a,b) several unconformities and multiple times of tect onism are developed in the Cretaceous to Lower Tertiary interval, resulting in a single unconformit y (Fig. 3.15). During deposition of Chorro Gp (La Paz and Esmeraldas Fms) there was fault activity (Ecopetrol et al., 1994). Topographic relief on the unconformity is up to 920 m near the persistent Cáchira (N-S oriented , Fig. 3.7) and Chucurí-Casabe (SW-NE oriented) palaeo-highs. The palaeo-highs separate two sub-basins to the northwest (Cantagallo-Yarí area). During deposition of the Chorro Gp. the irregular basin began to fill. The southwest end of the Chucurí-Casabe palaeo-high was an erosional highland while the Cáchira palaeohigh was overlapped (Cross et al., 1996a,b). The first phase of basin filling is rich in sandstone. Multiple episodes of sub-basin filling are revealed in the basal onlap pattern. The top of Chorro Gp (approximately Orange horizon of Cross et al., 1996a,b) is an onlap surface for the overlying unit. The upper boundary of the interval is characterized by minor erosion (Cross et al., 1996a,b). NW
Seismic Profile ME94-1400
SE . F m 0 a r a l c a u Le o g n F A m
m. F n L e o
m. F e r a n o b C a r
.
C a r b o n e r a
1.0
F m .
2.0 ) c e s M ir a d ( o r F m T . W T 3.0
F m. M i r a d o r
K
K
4.0
0
5 km
Figure 3.14: Seismic profile in the Medina area, eastern flank of the EC. Note contractional inversion of Cretaceous extensional f aults occurred during Palaeogene time (from Linares, 1996). Location of this profile is indicated in figure 3.2.
89
Chapter 3
During deposition of the Mugrosa Fm, continued onlap occurred onto the Chucurí-Casabe palaeo-high, which remained as a palaeo-high during deposition of the lower interval (Orange-purple of Cross et al., 1996a,b). However, it almost disappeared at the end (a more evenly thick isopach of the upper interval) during a gradual elimination of topographic relief. Each interval represents a basal fill episode, limited by an erosion surface and bi-directional onlap on that surface. Minor episodes of tectonic movement were recognized on some basal onlap patterns. Accommodation space/sediment supply ratio increased within each interval and between different intervals (Cross et al., 1996a,b).
m K 1
D
C
B
A
4 a o y a P
E
T L U A F A IN L A S A L
D
4 s a r a l C s a u g A
T B
B
W N
C
E
D
D
C
C 0
F
T L U A F A O Y A P
1 2 a o y a P
1 s a r a l C s a u g A
E
0 . 1
0 . 2
) c e s ( T W T
90
0 . 3
e l n t o r a a i e t t L a a . m B l e r . t o e o f e n N d e c . s o t u i s u r o M h n o o t r t k h c y c a r a B n y n . s r e T g t n a B i u . c . n Q s . e 2 . u 3 o d A e i e v . c a e r y u t e e g l i l r m f e a C n t i V e h t e n d a n e e L e e t l a . t a w c i d E e d g . b a s s i n M u e i s o i t e e i s l c e d a m l r d e i t f o i r f M C n o r p o e t s h e t t e c n s e u n a i l h r t a e l f n e n u o i e l g n c c n i o o t a i e d t n l a n a A P c o a . a ó L s e y . D ) a . c 2 n P e 8 e e n 9 u e h 1 q t c e , . o s h E e l g a u t o n o e e r t g s h e t n o a e v s e a o e c l l o a i K f g P i o l f r p O o m o r f c l y s i ( s e r m n n a k s E o i t e . i c i a S C t h : t 5 e . f n e 1 o i . m n 3 e s d e c e e g r o s n u g i a d g l i h n F O c a
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes
m K 1 A
B
C
D
E
6 2 a o y a P
T O A L Y U P A F A
C
B
T B
C
d e u n i t n o C : 5 1 . 3 e r u i F
T L U A F A IN L A S L A
D
W N 0 . 0
0 . 1
0 . 2
) c e s ( T W T
91
0 . 3
Chapter 3
During deposition of the Colorado Fm local tectonic reactivation of the Cáchira palaeo-high occurred and remained prominent. The lower boundary is a very well marked erosion surface that cuts into the underlying interval and is characterized by a pronounced facies offset. Cross et al., (1996a,b) recognized three intervals, each one characterized by highly aggradational vertically stacked mud-rich facies associations culminating in shales intervals deposited during periods of maximum base-level. Accommodation space/sediment supply ratio increased within each interval and between different intervals (Cross et al., 1996a,b). Detrital source areas. Sandstone of the La Paz, Esmeraldas, Mugrosa and Colorado Fms are characterized by quartz, sedimentary chert and K-feldspar as principal detrital components, with more K-feldspar than plagioclase and accessory aplitic lithic fragments, micas and zircon. An upward trend of increasing mineralogical maturity is observed (Hathon and Espejo, 1997). These sandstones fit into the craton/continental block provenance field on quartz, feldspar and lithic fragments plots (Dickinson, 1985). Hathon and Espejo (1997) interpreted an exhumated plutonic/high-grade metamorphic basement as source for the sediments, probably located in the area of the Central Cordillera ( Figs. 3.10 a nd 3.12). However, southeasterly and easterly local sources from the area of the EC are suggest ed by palaeocurrent measurements and grain size trends in the Opón, Malpaso and Río Sogamoso areas (Hathon and Espejo, 1997). According to Geotec (1994) dramatic lateral changes of thickness and palaeo-current with a westward dominant direction during deposition of the Esm eraldas Fm suggest a local detrital source area located probably in the Santander Massif (Figs. 3.10 and 3.12). Evidence presented here indicates that some parts of the EC started to be exhum ated and to sup ply sediment, thus demonstrating that the late pre-Andean foreland basin proposed by Cooper et al. (1995) was segmented by local exhumated areas. Upper M agdalena Val ley and H onda sub-basin
The Gualanday Gp of the Upper MV and similar coarse detrital units of the Honda sub-basin, i.e. the Hoyón and San Juan de Rioseco Fms. All these rock units are characterized by the presence of conglomerate and conglomeratic sandstone. The base of the Gualanday Gp varies spatially from conformable to angular unconf ormable (Figs. 3.8 and 3.9). This geometric variability, along with strong thickness and facies variatio ns ( Figs. 3.10 to 3.13), suggests both palaeogeographic and tectonic co ntrols on its accumulation. In the Gualanday Gp in the Neiva sub-basin (southern part of the Upper M V, Fig 2.1), Anderson (1970, 1972) described three conglomerate intervals separated by two mottled red mudstones and lithic sandstone intervals. Conglomerates are predominantly close-packed made mainly of chert pebbles. Conglomerates and sandstones show an upward increase in quartz through the section. The lower contact of the Gualanday Gp in the Neiva sub-basin has been reported to be paraconfo rmable, at least locally. However, in the Girardot sub-basin (northern part of the Upper MV, Fig 2.1) various levels of the Gualanday Gp overlap tilted and deformed Cretaceous rocks or older rocks (Figs. 3.8 and 3.9, Raasvelt, 1956; Raasvelt et al., 1957; Guillande, 1988; Amaya and Santamaría, 1994; Etayo- Serna, 1994; Ojeda and Peña, 1994). Guillande (1988) also reported local unconformities at the top of Potrerillo and Doima Fms of the Gulandaly Gp. The conglomeratic levels were deposited in alluvial fans and braided streams, while the red mudstone or siltstone levels suggest deposition on flood plains between channels. Coarse detrital fluvial/alluvial fan facies are overlain by lacustrine (?) mud–rich facies (Santa Teresa Fm. in the Honda sub-basin, and Barzaloza Fm in the Upper MV) suggesting a relative base-level rise.
92
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes Detrital source areas. Palaeo-currents
and clast composition indicate that the gra vels were derived for m Cretaceous strata, mainly due to episodic exhumation of the Central Cordillera (Figs. 3.10 and 3.12, Beltrán and Gallo, 1968; Van Houten and Travis, 1968; Anderson, 1970, 1972). C hert clasts were derive d from erosion of Upper Cretaceous rocks while detrital kaolinite-rich mudstones probably were derived from Cretaceous shales (Anderson, 1970, 1972). With increasing depth of erosion through time the quartzose lower part of the Cretaceous supplied an increasing amount of quartz to the basin. Van Houten and Travis (1968) postulate two pulses of exhumation of the Central Cordillera, the first represented by the lower Gualanday Gp (Chicoral Fm), while the second episode corresponds to the uppermost part of the Gp (Doima Fm). Anderson (1970, 1972) postulated three pulses of exhumation, one for each of the three conglomeratic levels described by him in the Gualanday Gp. Active thrusting during Eocene-Oligocene times has been proposed in the Upper MV (Butler and Shammel, 1988; Guillande, 1988; Caicedo and Roncancio, 1994; Amezquita and Montes, 1994; Amaya and Santamaría, 1994). Locally in the western part of the Girardot sub-basin the Gulanday Gp has been related to exhumation along the Chaparrral thrust fault (Caicedo and Roncancio, 1994), close to the western border with the Central Cordillera. They demonstrated that deposition of the Gualanday Gp occurred simultaneously with east-vergent thrusting and that sediments were derived from erosion of progressively deeper stratigraphic levels, including a Cretaceous sedimentary cover, Jurassic volcanic rocks and Palaeozoic igneous and metamorphic basement.
4.
SUBSIDENCE ANALYSIS OF THE PALAEOGENE SEDIMENTARY RECORD
The study of the sedimentary fill of a basin and its associated tectonic subsidence signal gives important information about basin formation mechanisms (Cloetingh et al., 1993). In order to quantify the tectonic component of subsidence of the Palaeogene basin of the study area, a 1D-backstripping technique was used (Steckler and Watts, 1978; Bond and Kominz, 1984). Details of the backstripping subsidence analysis, the procedure and data are described in Chapter 2. Figure 3.2 s hows the location of the columns and wells used. The Palaeogene sedimentary record of the study area only has been preserved in the LLA and MV areas, and partially along the axial zone of the EC ( Fig. 2.1, Sabana de Bogotá, Tunja and Sogamoso regions). In these regions only a very limited number of stratigra phic sections can be used for the subsidence analysis. The original data from the stratigraphic columns have been interpolated or extrapolated assuming the interpretations dr awn on the palaeogeographic and thickness maps were vali d (Figs. 3.3 to 3.5 and Figs. 3.10 to 3.13). Where possible I used also the restored original thicknesse s estimated fr om ther mal organic matter maturity modelling calibrated by vitrinite reflectance data from the available literature (Keal, 1985; Hébrard et al., 1987a,b; Bachu et al., 1995; Cazier et al., 1995; Mayorga and Vargas, 1995; Ecopetrol and Beicip, 1995; García-Gonzalez et al., 1997; Mora, 1997, Mora et al., 1997).
4.1.
RESULTS
Figure 3.16 shows some examples of the tectonic subsidence curves during early Tertiary times. Magdalena Valley.
Where the sedimentary record has not been partially eroded, or it is incomplete, there are important differences in the shape of the subsidence curves for different wells. (1) Middle Magdalena Valley Detailed thickness maps of several stratigraphical intervals in the Middle MV (Ecopetrol-ICP, 1996) show dramatic lateral changes in thickness, which together with evidence for active tectoni cs (Fig. 3.15) suggest piggy-back sedimentation. In the Guaduas Syncline
93
Chapter 3
(Honda sub-basin, Fig. 2.1) Gomez (1999) have shown seismic evidence of Palaeogene sedimentation simultaneous with folding. Maximum tectonic subsidence occurred in the eastern border of the MV (e.g. Tablazo section, Palaeogene tectonic subsidence 500 m) close to the faults limiting the western border of the EC. The convex upward shape of the tectonic subsidence curves indicating increasing subsidence rate through time during Palaeog ene (Fig. 3.16) is similar to the foreland basin subsidence pattern (Allen and Allen, 1990). (2) Upper Magdalena Valley. Amount of Palaeogene tectonic subsidence in this area is variable, but less than 500 m. Some authors (Amezquita and Montes, 1994; Amaya and Santamaría, 1994; Caicedo and Roncancio, 1994) have demonstrated Palaeogene sedimentation coeval with both east- and west-vergent thrusti ng (Etayo-Serna and Florez, 1994). Detailed thickness maps and local evidence (Figs. 3.8 a nd 3.9) also suggest piggy-back sedimentation. Llanos Orientales. The convex upward shape of tectonic subsidence curves ( Fig. 3.16) clearly shows acceleration of tectonic subsidence in a typical foreland pattern (Allen and Allen, 1990). Maximum Palaeogene tectonic subsidence is about 250-m in the Arauca–1 w ell (Fig. 3.16b). Sabana de Bogotá. A limited number of stratigraphic columns represent an incom plete sedimentary record. The most complete stratigraphic section in the Usme Syncline (La Calera and Cáqueza curve with about 750 m of tectonic subsidence, suggests a relative rapid Paleocene subsidence during deposition of the Guaduas and Bogotá Fms. Julivert (1970) has presented evidence of Palaeogene folding simultaneous with sedimentation in the Usme Syncline. Tunja-Sogamoso region. Limited data suggest two periods of relative rapid subsidence during the Paleocene (Guaduas and Socha Fms) and the Oligocene (Concentración Fm). Maximum tectonic subsidence is about 400 m. 4.1.1. Events of tectonic subsidence
Although there are important differences in the subsidence pattern of different stratigraphic columns the following subsidence events were identified: (1) Late Maastrichtian-Paleocene accelerated subsidence, especially in the axial zone of the palaeo-EC (Sabana de Bogotá region). (2) The Eocene unconformity is represented as a period of no subsidence. (3) Oligocene increased subsidence is regionally recognized. The following is a preliminary attempt to correlate events observed in the subsidence curves with plate tectonic events as proposed in the literature: During Maastrichtian and Paleocene several authors (McCourt et al., 1984; Pindell and Tabbutt, 1995) have interpreted accretion of the oceanic terranes of the Western Cordillera, with maximum tectonic activity during the Paleocene. This accretion produced exhumation in the palaeo-Central Cordillera, development of east-vergent thrusts in the Middle MV, and initial pre-And ean deformation and differential rock-uplift/subsidence of blocks in the area of the EC. As mentioned i n Chapter 2 i ncreased subsidence in the axis of the Cundinamarca sub-basin (Sabana de Bogotá) c ould be the r esult of increased horizontal compressional stress (Cloetingh, 1988; Cloetingh and Kooi, 1992) associated with collision of the oceanic terranes of western Colombia and deformation and uplift of the Central Cordillera. Development of normal faults in the Llanos area ( e.g.K luth et al. 1997) could be the result of local tensional stresses in the flexural bulge developed.
94
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes
MIDDLE MAGDALENA VALLEY 200
150
100
0
50 o e s l a a Eocene a P
Age (Ma)
M
EL
0
e e n n e e c c o o i g M i l O E
200 400 600 800 1000 1200 1400 1600 1800
W Lebrija
) m ( e c n e d i s b u S c i n o t c e T
Llanito-1 Sabalo-1k Casabe-199 Infantas-1613 Cascajales-1
2000
EASTERN CORDILLERA 250
200
150
100
50 -4000
e e o n n s l e e a a Eocene e c c o o i a P g M i M l E L O E
-3000 -2000
Age (Ma)
-1000 0 1000 2000 3000
) m ( k c o r f o f i l p U e c n e d i s b u S c i n o t c e T
Arcabuco Los Medios Pajarito Tibasosa Velez Lebrija
LLANOS ORIENTALES 100
80
60
40
20
0
0 200
Age (Ma)
400 600 800 1000
s Paleocene a a M E
e n e c o g i l O
Eocene
L
1200 e y n l r e a c o i E M
1400
) m ( e c n e d i s b u S c i n o t c e T
La Maria-1 Arauca-1 Caño Limon-1 La Tortuga-1
1600
Figure 3.16: Curves of tectonic subsidence of some wells (s ee Fig. 3.2 fo r location). Horizontal axis represents age in Ma. Vertical axis represents tectonic subsidence in meters o btained fro m backstripping analysis.
95
Chapter 3
The Eocene unconformity correlates with an increase in convergence rate between Caribbean and South American plates. During this time a maximum peak of deformation occurred (first event of pre-Andean deformation). During Oligocene time a second event of pre-Andean deformation occurred in the EC. It correlates with thin-skin obduction of Panamá (Chocó block) onto Western Cordillera-San Jacinto deformed belt (Pindell and Tabbutt, 1995), and with the break-up of Farallon plate into the Cocos and Nazca Plates (Wortel and Cloetingh, 1981; Duncan and Hardgraves, 1984). 4.1.2. M aps of tectonic subsidence
Figures 3.17 t o 3.20 sho w maps of tectonic subsidence for selected time intervals during the Palaeogene. Maximum tectonic subsidence during the Paleocene occurred in the Sabana de Bogotá area (La Calera, Cáqueza) as indicated in the thickness map. The calculated tectonic subsidence for the Eocene in the EC is very small and the only significant tectonic subsidence occurred in the MV, as suggested by the Eocene thickness map. The Oligocene to Early Miocene subsidence was maximum in the eastern and western foothills of the EC.
Serrania de Perija
Maracaibo Basin
1400
e s d A n a r i d e M
Cucuta s a c u L n a S e d ia n a r r e S
1300 0.00
Medellin
1200
Manizales
8.00
124.00
0.00 1.00
20.00 36.00
Ibague
62.00
50
24.00 29.00
128.00
36.00
t l u a f o52.00 e a l a p 45.00 a t i v r e S57.00
100
0.00
0.00
39.00
0.00
21.00
106.00
0.00
0.00
0.00
0.00
0.00
0.00
100.00
0.00 0.00
198.00
0.00 0.00 0.00
Villavicencio 0.00
10.00
0.00
49.00
0.00
2.00
l t ra e n C 800
800
0.00
0.00
0.00
0.00
0.00 81.00
Neiva 0.00
0.00
900
Llanos Orientales Basin (LLA)
0.00
0.00 0.00
Bogota
200 Km
Arauca
0.00
c a u l t y a f a o B e o 90.00 l a p a
339.00
0.00
900
0
33.00
108.00
106.00
18.00 14.00 19.00
26.00
157.00 82.00
109.00
9.00
0.00
m e m t e 110.00 s t 0.00 95.00 s 82.00 y y 79.00 103.00 s t s 78.00 u l t 123.00 0.00 a l 82.00 f 144.00 u 0.00 o 147.00 f a 8.00 Tunja 129.00 a e l 98.00 169.00 o 111.00 p a 0.00 119.00 a e 11.00 l o Yopal 0.00 a 105.00 m p r a a 233.00 c á0.00 69.00 I m 86.00 i 129.00 0.00 u a u t G0.00 i 84.00 B 7.00
0.00
1000
44.00
Bucaramanga
r a i le d o r C
0.00
1100
34.00
Barinas Basin
a e n e d r a a i n c a a M r r e a S L
1000
0.00
San Jose del Guaviare
1100
Guyana Shield
1200
1300
BACKSTRIPPED LATE MAASTRICHTIAN EARLY PALEOCENE TECTONIC SUBSIDENCE
1400
Figure 3.17: Observed late Maastrichtian-early Paleocene tectonic subsidence, in meters, calculated from backstripping analysis.
96
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes Serrania de Perija
1400
s a c u L n a S e d i a n a rr 0.00 e S
1300
Maracaibo Basin
105.00
-1.00 22.00 117.00
Medellin
1100
0.00
Manizales
20.00
27.00
1000
19.00
Ibague
17.00 23.00
29.00
140.00
385.00
88.00
142.00
900
800
800
123.00
23.00
25.00
Arauca
40.00
0.00
0.00
Llanos Orientales Basin (LLA)
38.00
0.00
0.00
0.00
0.00
0.00 0.00
10.00 26.00
0.00
52.00 14.00
6.00 32.00 5.00
0.00
0.00
18.00 0.00
0.00
30.00 0.00
0.00 11.00
0.00 0.00
Villavicencio
0.00 10.00
l m n t ra s y s t e e t C f a u l
e o l a a p a m u s C h 0.00
200 Km
0.00
90.00
e m s t y t s u l a f e o7.00 a l p a o m r a 42.00 0.00 c á i 53.00 u a G Yopal
0.00
8.00 0.00
39.00
243.00 0.00
0.00
114.00
62.00
Bogota
20.00
100
Bucaramanga
m 39.00 t e s 11.00 y -1.00 s 21.00 t 49.00 Tunja l 49.00 u 1.00 66.00 85.00 f a 4.00 19.00 o 68.00 17.00 a e l 57.00 a p 101.00 a 73.00 m 99.00 i 107.00 u t i 66.00 102.00 31.00 B
0.00
0
t 0.00 l u f a 130.00 o 102.00 t e l a l a19.00 a c a u a f p 32.00 o y o z 109.00 e B l a e 97.00 110.00 r a u a 16.00 p 155.00 S
r a i le d o r C
Barinas Basin
-
41.00 0.00
1200
e s d A n a r i d e M
Cucuta
14.00
73.00 0.00
BACKSTRIPPED LATE PALEOCENE TECTONIC SUBSIDENCE
0.00 0.00
70.00
0.00
0.00
Neiva
900
0.00
a e n d e r a a i n c a a r M r e a S L
1000
San Jose del Guaviare
0.00
1100
1200
1300
1400
1500
Figure 3.18: Observed late Paleocene tectonic subsidence, in meters, calculated from backstripping analysis.
5. MODELLING OF REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING Because lithosphere stretching generated an extensional basin during the Mesozoic, the first hypothesis tested to explain the Palaeogene observed tectonic subsidence, was whether some remaining thermal subsidence affected Palaeogene evolution ( c.f. Deségaulx et al., 1991). Assuming that the basin was not deformed and that the only basin subsidence mechanism during the Palaeogene was remaining thermal subsidence following Mesozoic rifting, I calculated for several time intervals the remaining thermal subsidence due to thermal re-equilibration of the lithosphere during Palaeogene. A description of the forwar d modelling technique and procedures developed by Van Wees et al. (1998) is given in Chapter 2. I used the lithosphere stretching factors calculated for Mesozoic extensional phases and all the modelling parameters used in the Mesozoic rifting episodes (see tables 2.4a,b in Chapter 2). Finally I compared the Palaeogene thermal subsidence with the observ ed tectonic su bsidence, previously calculated using backstriping techniques, in order to test the hypothesis of tectonic quiescence within the basin during the early Tertiary.
97
Chapter 3 Serrania de Perija
Maracaibo Basin
1400 s a c u L n a S e d i a n a rr e S
1300 0.00
Medellin
1200
0.00
1100
a l e r rdi o C
0.00
0.00
4.00 0.00
2.00
o e la e m a p st 4.00 a s y im l t 17.00 u it au B f
22.00
30.00 -1.00 0.00
Tunja 5.00
15.00
17.00 9.00
10.00
15.00
11.00
28.00
33.00
14.00
16.00
19.00 25.00 20.00 21.00
48.00
16.00 29.00 16.00
14.00
20.00
19.00
17.00 15.00
0.00 8.00 10.00
21.00 23.00
12.00 14.00
16.00
-2.00 24.00 35.00
34.00 47.00
17.00
1.00
39.00
0.00
3.00 10.00 0.00
21.00
25.00 12.00
79.00
28.00
11.00
29.00
20.00
0.00
3.00
0.00
191.00
4.00 1.00 0.00
106.00
0.00
a e n d e r a a c i n a a M r r e a S L
900
11.00
12.00
Neiva
800
Llanos Orientales Basin (LLA)
24.00
Yopal
32.00
0.00
Villavicencio
800
28.00
-1.00
17.00
-1.00
Bogota
76.00
200 Km
Arauca
24.00
21.00 22.00
55.00
l t ra e n C
22.00
18.00
3.00
5.00
41.00
11.00
14.00
4.00
36.00
72.00
132.00
Ibague
-1.00
-2.00
100
3.00 24.00
27.00 0.00
52.00
0
5.00
-2.00
1000
l a e p a
Barinas Basin
-
t l Bucaramanga 65.00 u a f o e t l 0.00 a 10.00 l u a 2.00 a 98.00 f p 18.00 a e o t i a l 3.00 v r a p e S z 10.00 e a t r l a 15.00 a c a u u S 11.00 B o y o f 21.00
9.00
0.00
8.00
0.00
66.00
106.00
Manizales
900
e s d A n a r i d e M
Cucuta
1000
0.00
BACKSTRIPPED EOCENE TECTONIC Guayana SUBSIDENCE Sheld
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 3.19: Observed Eocene tectonic subsidence, in meters, calculated from backstripping analysis.
5.1. RESULTS Figures 3.21 to 3.24 sh ow the theoretical remaining thermal subsidence in a scenario of te ctonic quiescence. A comparison of these theoretical remaining ther mal subsidence maps ( Figs. 3.21 to 3.24) with the observed tectonic subsidence (Figs. 3.17 t o 3.20), clearly shows profound differences im plying that the hypothesis of tectonic quiesce nce is not ap plicable to the Palaeogene history of the EC. In general in the place of the former extensional basin (palaeo-EC) the observed tectonic subsidence is smaller that the predicted thermal subsidence. Thus if a component of remaining thermal subsidence occurred after the Mesozoic rifting, such subsidence was partially compensated by a moderate relative rock-uplift produced by a different tectonic mechanism. Conversely in the MV and LLA foothills areas, just out the margins of the former ex tensional basin, the observed tectonic subsidence is larger than the theoretical rem aining thermal subsidence, implying an additional subsidence produced by some tectonic mechanism ( Figs 3.17 t o 3.24). These trends can be explained at least qualitatively by assuming that an early e pisode of inversion of the former extensional basin occurred during Palaeogene time.
98
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera, Colombian Andes
s a c u L n a S e d ia n a r r e S
1400
ra l l e i rd o C
1300
Maracaibo Basin
Serrania de Perija
40.
0. 169.
100
103.
0.
187.
303.
42.
199.
201.
145.
164.
0.
212.
0.
0. 180.
25.
25.
97.
110.
107.
Tunja
7.
264. 154. 288. 359.
98.
259.
374.
207.
260.
356.
210.
93.
135. 156.
163. 170.
160.
148.
145. 139.
165.
218. 256.
178.
146. 156.
261.
442.
127.
265.
240.
245.
68. 114.
187.
182.
Yopal
186.
131.
129. 123.
119.
374.
Villavicencio
45 7.
177.
166.
132.
34.
269.
261.
162.
297. 217.
113.
124.
112.
179.
310. 178.
BACKSTRIPPED OLIGOCENE EARLY MIOCENE TECTONIC SUBSIDENCE
177.
800
114. 137.
Neiva
a e n e d r a a i n c a a M r r e a S L
11.
800
900
1000
36.
184.
338.
l n t r a e C
Llanos Orientales Basin (LLA)
285.
87.
Bogota
Ibague
119.
179.
177.
6.
37.
148. 116. 33.
172. 231.
26.
175. 181.
Manizales
22.
6.
0. 43.
50.
104.
0.
139.
119.
49.
0.
32.
Arauca
23.
0.
18.
200 Km
158.
0.
900
0
186.
Medellin
1000
Barinas Basin
-
Bucaramanga
1200
1100
e s d A n i d a r M e
Cucuta
111.
San Jose del Guaviare
1100
Guayana Sheld 1200
1300
1400
1500
Figure 3.20: Observed Oligocene-Early Miocene tectonic subsidence, in meters, calculated from backstripping analysis. Negative values mean tectonic uplift.
6. FLEXURAL BEHAVIOUR OF THE LITHOSPHERE During processes of mountain building, the lithosphere bends due to the topographic loading (Price, 1973). This bending demonstrates that the lithosphere is capable of supporting localized loads such as mountain ranges by deforming over a much broader area. As a result, depressions are formed at both sides of a mountain chain in which sediments can accumulate. The origin of these foreland basins is, therefore, controlled mainly by the flexural behavior of the lithosphere (Beaumont, 1981). Flexural studies that treat the lithosphere as an elastic sheet floating on a fluid asthenosphere simulates the observations (Beaumont, 1981). Flexural studies of the lithosphere are constrained by the geometry of the basement subsiding asymmetrically under the load of the mountain belt ( e.g. Zoetemeijer, 1993). Not only surface mass loads, but also intraplate forces can play a significant role in the development of foreland basins (Cloetingh, 1988). In some flexural studies the topographic load has proved not sufficient to explain the observed deflexion of the foreland ( e.g.R oyden and Karner, 1984).
99
Chapter 3 s a c u L n a S e d ia n a r r e S 6
1400
1300
Maracaibo Basin
Serrania de Perija
e s d A n i d a r M e
Cucuta
50
Barinas Basin
60 7 2
29 16 12
24
-
0
100
200 Km
Bucaramanga
18
22
16
0
20 27
18
0
Arauca
39
15
0
51
0
41 53
Medellin
1200
35
32 24
a l e r i o rd C
20
26
15
40
8
50
15 32 51
47
0 32
30
58
23
Ibague 14
27 19
20
8
22
23
17
32 39 41
41
44
40
27
53
0
Villavicencio
30
0
0
0
0
0
0 0
0
0
0
32
30
0
0
0 0
31
17
0
0
0
0
0
0 0 0
PREDICTED LATE MAASTRICHTIAN EARLY PALEOCENE THERMAL SUBSIDENCE (m)
22
Neiva a e n d e r a a c i n a a M r r e a S L
20 1 7
800
0
0 0
0
25
800
0 0
30
31 20
00 0
0 0 00
0
0
0 0 0 0 0
0 30
0 0
Yopal
0
40
Bogota
35
0 0
Llanos Orientales Basin (LLA)
0
0
31
43
43
l t ra e n C
0
34
30
0
0
40
33
50
900
35
Tunja 38
30
0
40
27 16 42
19
26
37
Manizales
1000
48
11
74
1100
0
14
900
1000
0
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 3.21: Predicted late Maastrichtian-early Paleocene thermal subsidence af ter Mesozoic rifting, in meters, calculated from forward modelling using lithosphere-stretching factors discussed in Chapter 2. Negative values mean tectonic rock-uplift.
In these cases the introduction of plate boundary forces produces better model results but the physical meaning of these forces is still a matter of debate (Zoetemeijer, 1993). The strength of the lithosphere is an important factor in the development of foreland basins. The distribution of strength in the lithosphere varies vertically as a result of its thermal and compositional layered structure. Application of a depth-dependent rheological model to the continental lithosphere implies stacking of two strong elastic layers, embedded in weak layers which yield by brittle or ductile deformation (Burov and Diament, 1995; Cloetingh and Burov, 1996). Such a model represents more accurately the behaviour of continental lithosphere, which is necessary for the adequate analysis of strength and stress distribution (Zoetemeijer, 1993). However, an appropriate first order description of the flexural behaviour of the lithosphere can be formulated assuming that it behaves as a thin elastic plate ( e.g.T urcotte and Schubert, 1982). An important independent constraint in flexural studies is the gravity field (Karner and Watts, 1983; Zoetemeijer, 1993).
100
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes Serrania de Perija
1400
Cucuta s a c u L n a S e d ia n a r r e S
1300 -19
1200
Medellin
-30
0
8
14 20
11 29
36 19
20 30
43
11
1 2
13
Tunja 32 30 35 26
22
11 22
33
14
0
30 23
31
0
Yopal
40
21
0 0
0 0
0 0
0
0 0
0
30
0
0
0
0
0 0
34
30 23
0
Villavicencio
0
0
0
0
800
900
100
200 Km
0 0
0
20
14
0
0
24
Neiva
0
0
PREDICTED LATE PALEOCENE THERMAL SUBSIDENCE (m)
0
19
800
0
0 0
0
30
0
0
0
25
0 0
0
0
0
0 0
13 30
Llanos Orientales Basin (LLA)
0
0
5
16
30
16 14
0
33
20
25
0
40
23
Bogota
28
19
25
1 1 25 3 7
26
25
34
19
7
16
Manizales
l n t ra e C
0
20
7
4022
17
Arauca
0
3 9 39
a le r i o rd C
Ibague
0
29
23 17
7
1000
50
Bucaramanga
12
60
1100
Barinas Basin
54
23
11 10
4 13
16
900
e s d A n a i d e r M
Maracaibo Basin
0
a e n d e r a a c i n a a M r r e a S L
San Jose del Guaviare
1000
1100
1200
1300
1400
1500
Figure 3.22: Predicted Late Paleocene thermal subsidence after Mesozoic rifting, in meters, calculated from forward modelling using lithosphere-stretching factors discussed in Chapter 2. Negative values mean tectonic rock-uplift.
6.1. MODEL DESCRIPTION The flexural response of the lithosphere to applied loads is a function of the vertical load due to a mass distribution, the horizontal force and the effective elastic thickness (EET) of the assumed thin elastic plate, which can change laterally. The parameter that characterizes the apparent strength of the lithosphere is the flexural rigidity D, which is commonly expressed through the EET of the lithosphere (Zoetemeijer, 1993). A useful parameter to judge how intense the flexural deformation of the lithosphere is and whether the assumption of elastic deformation is still valid, is the flexural bending stress σ xx or fiber stress (Ranalli, 1995), which for a thin elastic plate is given by
σ xx
Ey = − 2 1 − ν
101
d 2 w dx 2
(3.2)
Chapter 3
Where y represents the vertical axis, x represents the horizontal axis along the profile, E r epresents the Young’s modulus, ν represents the Po isson’s ratio and w the deflection. A first step in extending the model fr om pure elastic deformation ( Figure 3.25a) to depth-dependent multilayered decoupled rheology ( Figure 3.25b) is setting a n upper bound for the applied elastic deformation (following Turcotte and Schubert, 1982). Assuming an elastic-plastic rheology for the bending of the plat e (Figure 3.25c), the stress increases linearly with distance y (in depth) from the center of the plat e (y = 0). The plate bends elastically until the stress at the surface becomes sufficiently large that yielding occurs. The yield stress may be simplified to σ* = σ xx.. From these relations and using the expression for the bending moment M(x) = -D(x)d 2w/dx2 , one can determine the bending moment corresponding to the onset of plasticity M on for y = ½h (h=EET) and M c the critical bending moment for the situation of entire yielding for y=0: M on = - σ* h 2 /6 and M c= - σ* h 2 /4 (Zoetemeijer et al., 1999; for details see Turcotte and Schubert, 1982, section 7.11). All flexural models in this chapter gave fiber stress values well below the onset of plastic yielding considered as 1GPa. following Zoetemeijer et al. (1999). Serrania de Perija
1400
Maracaibo Basin Cucuta
s a c u L n a S e d i a n a r r e S
1300 65
Medellin
1200
103
47
13
23
139
0
28
34 51
30
4 0
0 94
31
29
79 75 91
40
60
66
0 0
76
41
Ibague 27
80
51
41
34
80 60
53
54
103
86
40
0
Villavicencio
0 0
0 0
60 0
20 35
800
900
1000
0
0
0
a e n d e r a a c i n a a M r r e a S L
200 Km
0 0
0
Neiva
10 0
0
43
800
0 0
0
71
0
0
0
0
0
61
00 0
0 0 0
0
0 0
0 0 0 0
0
80
0
0
0 0 0 0
60
60
0
0 0
32 74
Bogota
67
15
39
57
l n t r a e C
0
0 Yopal
60
34
86
46 35
0
68 113
1000
0
80
54
Manizales
Llanos 0 Orientales Basin (LLA)
0
99
62
Tunja
35
67
28
29
0
80
52
35
0
102
48
100
23
Arauca
0
10 0
77
12
44
0
75
15
40 60 80
Barinas Basin
Bucaramanga
34
57 44
r a i le d o r C
100 120
59
21
144
1100
900
e s d A n a r i d e M
0
San Jose del Guaviare
1100
1200
PREDICTED EOCENE THERMAL SUBSIDENCE (m)
1300
1400
1500
Figure 3.23: Predicted Eocene thermal subsidence a fter Mesozoic r ifting, in meters, calculated from forward modelling using lithosphere-stretching factors discussed in Chapter 2. Negative values mean tectonic rock-uplift.
102
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
s a c u L n a S e d ia n a r r e S
1400
1300
30 40
16
7
Barinas Basin 200 Km
100
0
76
32
Bucaramanga
-229
50
16
-54
d a i r M e
Cucuta
-153
s d e n A
Maracaibo Basin
Serrania de Perija
8
18 2 7
1 6
0
0
Arauca
41
0
0
54
0
42 55
Medellin
1200
-85
31 24
r a i le d o r C
50
30
25
0 -2 15
-19
36
15
15
17
34
43
13 12
0
28
4
-526
1100
0
-195
Tunja 40
10
0
49
-194
Ibague 15 265
-123
0
27
0
0 0
17
29
0
0
0 0 0
0 0 0
0
5 6
0
0
0
0
0
32
0
900
l n t ra e C
0
10
0
0
0
0 0
0
0 0
18
800
900
a e n e d r a i a n c a a r M r e a S L
1000
0
0
-343 -157
0
0
0 0
-125
Neiva
0
0
Villavicencio
26
800
0
0 0
-55
0
0
47
18
-152
0
0 0
0
0 0 00
Bogota
-225
24
8
20
41 37
-287
0
0
Yopal
32
29
13
1000
0
0
36
61
Llanos Orientales Basin (LLA)
33
23
Manizales
0
53
San Jose del Guaviare
1100
0
PREDICTED OLIGOCENE EARLY MIOCENE THERMAL SUBSIDENCE (m)
Guayana Shield
1200
1300
1400
1500
Figure 3.24: Predicted Oligocene-Early Miocene thermal subsidence after Mesozoic rifting, in meters, calculated from forward modelling using lithosphere-stretching factors discussed in Chapter 2. Negative values mean tectonic rock-uplift
7. MODELLING OF FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PALAEO-CENTRAL CORDILLERA DURING PALAEOGENE TIME In this section, I test the currently accepted hypothesis proposed in literature ( e.g.C ooper et al., 1995) that the area of the MV, palaeo-EC and LLA was a simple foreland basin related to lithosphere flexure, produced by a topographic load represented by the palaeo-Central Cordillera. I tested this hypothesis through modelling lithosphere flexure, assuming the lithosphere behaves as an elastic plate with lateral variations in thickness (EET) and applying a load representing the topography of the Palaeogene Central Cordillera. I assumed a broken plate with a free boundary located at the end of the continental plate during Palaeogene along the Romeral suture just west of the Central Cordillera (Figure 3.2). I modelled lithosphere flexure along four 2D regional sections and also in 3D.
7.1. 2D FLEXURAL MODELLING I modelled the flexural lithosphere deflection in 2D sections using the program Cobra developed by Zoetemeijer (1998). The program calculates 2D deflections for distributed loads on a
103
Chapter 3
plate with a variable thickness. The 4 th order differential equation of the flexural behaviour is solved using a finite difference technique (Bodine, 1981). Input data for the model are: the length of the plate, the size of the finite difference cell, the horizontal co-ordinate of the plate boundary in case of a broken plate, the horizontal co-ordinate of the boundary between the load and the foreland basin, densities of the crust, mantle, mountain load and sediment filling the foreland basin ( Table 3.1). Several parameters can vary laterally: Elastic thickness of the plate, topographic load, additional surface or subsurface loads, initial and actual water depth of the foreland basin, and initial crustal thickness. As an independent source of information, gravity anomaly calculations help to obtain a more reliable model. For that reason the possibility to calculate the contribution to the gravity field is included in the program. PERFECTLY ELASTIC RHEOLOGY Tension Stress
0
Compression Stress
BRITTLE ELASTIC AND DUCTILE RHEOLOGY Tension Stress
h t p e d
Tension Stress
Compression Stress
A
ELASTIC PERFECTLY PLASTIC RHEOLOGY 0
0
Moho
Compression Stress
h t p e d
h t p e d
*
B
C
Figure 3.25: Cartoon showing different rheological models of continental lithosphere. (a) Perfectly elastic behaviour. (b) Depth-dependent brittle-elastic-ductile rheology. (c) Elastic-perfectly-plastic rheology of simplified continental lithosphere. (from Zoetemeijer et al., 1999).
Outputs of the program are calculated flexural deflection, gravity, and fiber stress produced by bending of the lithosphere plate, which can be displayed as profiles along the section. Table 3.1 shows assumptions and parameters I applied to model flexural subsidence due to topographic load of the Central Cordillera during Palaeogene time. Those unknown values (such as topographic elevation of the Central Cordillera during Palaeogene and elastic thickness) were modified several times by a trial and error procedure, until reaching the optimum fit between the calculated flexural deflection of the lithosphere and the observed subsidence during the Palaeogene. ASSUMPTIONS -Broken plate with a free boundary located westward on the Romeral suture zone ( Fig. 3). -No water depth (Palaeogene depositional environments were dominantly continental to transitional. -No horizontal intraplate force. -No vertical shear force at the free end of the plate. -No bending moment at the free end of the plate. -Sediment density applied as a load filling the basin when sediment thickness was used to compare to the calculated flexural deflection. -No sediment fill load applied when observed tectonic subsidence was used to compare to the calculated flexural deflection. -No subsurface loads, except for simulating thermal subsidence as explained in text. Table 3.1a. Assumptions used in lithosphere flexural models.
104
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes PARAMETERS -Length of the plate: 1000 km. -Size of the finite difference grid cell: 5 km. -Densities (kg/m3) -Sediments: 2250 -Topographic load: 2550 to 2650 -Crust: 2770 -Mantle: 3300 -Effective elastic thickness and topographic load variable as shown in Figs. 10 to 16a nd Table 2. Table 3.1b. Parameters used in lithosphere flexural models.
7.2. 3D FLEXURAL MODELLING I modelled the flexural lithosphere deflection in 3D using data displayed as maps with the program Flex3D developed by Van Wees (1993). The program Flex3D calculates 3D deflections for distributed loads on a plate with lateral changes in elastic thickness. The program was developed to forward model the vertical flexural effects of lithosphere stretching and shortening in three dimensions, focussing on flexural effects related to these processes. The forward model for stretching combines main features of two dimensional basin models developed by Kooi (1991) and Kusznir et al. (1991). Necking depths and variations in effective elastic thickness can be adopted. The differential equation for flexural behaviour is solved in 3D applying a finite difference formulation, which is solved by a direct matrix method. To incorporate the effect of spatial variations in rigidity, additional terms for the bi-harmonic 3D-flexure equation have been derived from a variational displacement formulation as used in finite-element methods. Additionally, planar faults are treated as discontinuities (Kusznir, 1991). The algorithm is described in Van Wees and Cloetingh (1994). Input data for the program are dimensions of the modelled plate, finite difference grid size. Variables varying spatially are given to the program as files include for example: effective elastic thickness, topography, palaeo-water depth and crustal thickness. Intraplate stresses can also be included in the program. I used the same parameters applied to 2D flexural models (Table 3.1).
7.3. RESULTS 7.3.1. 2D models
Initially I tested the present day topography of the Central Cordillera as if it had remained unchanged since Palaeogene times . Figure 3.26 sh ows the results of these models. Clearly the predicted subsidence is many times smaller than the observe d subsidence. Thus if the subsidence was produced by the Central Cordillera, a much higher topography than present day would be necessary. Additionally the shape of the calculated deflected profile is different to the observed profile of subsidence. In a second trial I assumed only a topographic load in the palaeo-Central Cordillera and I modified it and the effective elastic thickness until it produced the best fit with the observed subsidence. Figure 3.27 sho ws the results of these models. In all the cases the shape of the calculated subsidence profile is different to the shape of the observed subsidence profile. Only for the most eastern LLA area both profiles are similar. To produce the observed subsidence in the LLA area by flexural deflection, it would be necessary to apply 5000 m of topography for the palaeo-Central Cordillera and an elastic thickness of 200 km. These values are completely unrealistic: an elastic thickness of 200 km would imply a lithosphere several times stronger than stable cratonic lithosphere ( e.g. Cloetingh and Burov, 1996). These unrealistic results indicate that the observed Palaeogene subsidence in the palaeoEC and neighbouring MV and LLA areas cannot be explained only by flexural subsidence of the lithosphere due to the Palaeogene topography of the Central Cordillera. The observed Palaeogene subsidence in the Magdalena and LLA areas increase to a maximum close to the borders of the former extensional basin. Only a load located in the Central Cordillera could never produce this.
105
Chapter 3 e
The asymmetric profile of subsidence, with larger subsidence in the western Magdalena area e than in the eastern LLA area, probably was due to the load of the palaeo-Central Cordillera. However, in order to reproduce a deflection as the observed, it is necessary to adopt some additional load in the area of the former extensional basin. One possible load in such an area would be remaining thermal subsidence of the former extensional basin. However, we have already tested such a hypothesis (section 5 of this Chapter) and we found that thermal subsidence alone cannot explain the observed backstripped tectonic subsidence. An alternative hypothesis is to consider that the original extensional basin was partially inverted during Palaeogene, generating some amount of local topography that would produce flexural deflection in the lithosphere, with maximum values close to the borders of the e hypothesis through modelling. partially inverted extensional basin. I also tested this last
SECTION 1
SECTION 5
F LE X U R E
m
F LE X U R E 5000
5000
n i
0
0
ce l
e -5000
d
d
-10000
-10000
o
0
e
80
200
400
600
800
1000
1200
0
400
600
800
1000
1200
60 40
40
e
20
20
)
200
80 E F F E C T I V E EL A S T I C T H I C K N E S S
E F F E C T I VE E L A ST I C T H I C K N E S S
60
e
-5000
0
0 0
200
400
600
800
1000
0
1200
200
400
600
800
distance (km )
distance (km )
SECTION 3
SECTION 12
F LE X U R E
1000
1200
F LE X U R E 5000
50 0 0 0 -5000
-500
-10000 -1000 0.
200.
400.
600.
800.
0.
1000. 1200.
E F F E C T I V E E L AS T I C T H I C K N E SS
400.
600.
800.
1000. 1200.
80 EFFECT IVE ELASTIC THICKNESS
80
60
60
40
40
e
200 .
e
20 0 0.
200.
400.
600.
800.
1000. 1200.
20 0 0.
200.
400.
600.
800.
1000.
1200.
distance (km )
distance (km )
Figure 3.26: 2D lithosphere flexural models only assuming the present day topography load of the Central Cordillera as if it were remained unchanged since Palaeogene times. Top panel: basement deflection and topography in meters. Dots represent the observed thickness of Oligocene-Early Miocene deposits in meters. Bottom panel: effective elastic thickness in km. In order to try to fit the observed deflection and taking into account the post-rifting thermal weakening of the EC area reduced elastic thicknes s was assumed in this area. However in these models it was not possible to fit the observed deflection. Figure 3.2 s hows location of these sections.
106
t t
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes SECTION 1
SECTION 5
FLEXURE
m
FLEXURE
5000
5000
0
0
-5000
-5000
n o e c e e
0
500
1000
1500
0
EFFECTIVE ELASTIC THICKN ESS
)
20 0
m ( t
10 0
e
50
e
10 0 50 0
0
500
1000
1500
0
distance (km)
500
1000
1500
distance (km )
SECTION 3
SECTION 12
FLEXURE
FLEXURE 4000
5000
m
1500
15 0
e
0
)
1000
20 0
15 0
(
500
E F F E C T I VE E L A ST I C T H I C K N E S S
2000
n oi
0
t
0
c el
2000
-5000
f e d
0
500
1000
1500
0
E F F E C T I V E E L A ST I C T H IC K N E S S
e
20 0
20 0
15 0
15 0
10 0
10 0
50 0
e 0
500
1000
500
1000
1500
E F F E C T I V E E LA S T I C T H I C K N E S S
1500
distance (km )
50 0
0
500
1000
1500
distance (km )
Figure 3.27: 2D lithosphere flexural models assuming only a topographic load in the palaeo-Central Cordillera that best fits the observed thickness of Oligocene-Early Miocene deposits. Top panel: basement deflection and topography in meters. Dots represent the observed thickness of Oligocene-Early Miocene deposits in meters. Bottom panel: effective elastic thickness in km. Figure 3.2 sho ws location of these sections. 7.3.2. 3D m odel
As an additional test I modelled in 3D the flexural deflection of th e lithosphere produced by a topographic load in the Central Cordillera. Figure 3.28 shows the calculate d flexural deflection, which monotonously increases westward producing a deflection comparable to th e asymmetric regional long wave component of observed Palaeogene subsidence. However the model did not predict the observed local maximum subsidence in the former e xtensional basin and the LLA are a, nor the short wave length component of the observed Palaeogene sub sidence (e.g. compare Fig. 3.28 with Fig. 3.20).
107
Chapter 3
Serrania de Perija
s a c u L n a S e d ia n a rr e S
1400
1300
Maracaibo Basin
e s d A n a r i d e M
Cucuta
Barinas Basin
Bucaramanga Arauca
r a i le d o r C
Medellin
1200
Llanos Orientales Basin (LLA)
Tunja
1100 Yopal
Manizales
1000
0
Bogota
Ibague
100
200 Km
Villavicencio
900
l n t r a e C
LITHOSPHERE FLEXURE PRODUCED BY A UNIFORM LOAD ON THE CENTRAL CORDILLERA (m)
Neiva a e n d e r a i a n c a a r M r e a S L
800
800
900
1000
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 3.28: 3D lithosphere flexur al model. 3D f lexural deflection, in meters, of the lithosphere produced only by a topographic load located in the palaeo-Central Cordillera. See these results with thickness maps of Oligocene-Early Miocene deposits for a comparison (Fig. 3.13).
8.
MODELLING OF FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PARTIALLY INVERTED EXTENSIONAL BASIN DURING PALAEOGENE TIME
In this section I test the hypothesis of flexural subsidence produced by some local surfaceuplifted blocks (incipient basin inversion, Gomez et al., 1999) within the former extensional basin, as suggested by exhumation of some areas indicated by local sedimentological ( e.g. Anderson, 1970; 1972; Laverde, 1989; Ecopetrol-ICP, 1996; Gómez et al., 1999), petrographical ( e.g. Porta, 1966; Anderson, 1970, 1972), mineralogical (Cheilletz et al., 1993, 1997; Branquet et al., 1996), structural (e.g. Julivert, 1970; Butler and Schammel, 1988; Casero et al., 1995, 1997; Restrepo-Pace, 1999a,b; Gómez et al., 1999) and fission track data ( e.g. Shagam et al., 1984; Toro, 1990; Van der Wiel, 1991; Hossack et al., 1999; Gómez et al., 1999). I model flexural subsidence in 2D, along four regional sections, produced by only the topographic load of the partially inverted extensional basin during Palaeogene time using the same methods, parameters and software.
108
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
8.1.
RESULTS
Figure 3.29 sh ows that to generate a flexural deflection similar to the observed subsidence would require a topogr aphic load in the area of the former extensional basin. Such topographic load would be approximately 2000 m high, with effective elastic thickness of 60 km in the LLA areas, decreasing westward to values of approximately 20 to 25 km in the palaeo-EC and MV area. Although these values are theoretically possible ( e.g. Cloetingh and Burov, 1996), a topography of 2000 m for the palaeo-EC during Palaeogene time is in conflict with the presence of a Palaeogene sedimentary record in the axial zone of the Cordillera (Sabana de Bogotá, Tunja, and Sogamoso regions). The local evidence of marine sedimentation in the Oligocene Concentration Fm also contradicts this hypothesis. If such topography of 2000 m had existed, there would be abundant evidence of it in the sedimentary record, in the palaeocurrents pattern and detrital mineralogical composition (in the neighbouring LLA and Magdalena basins), that is not the case. Additionally many authors (Van der Hammen et al., 1973; Helmens, 1990; Hooghiemstra, 1984, 1989) have reported palynological evidence of the surface-uplift of the Sabana de Bogotá area in the last 5 Ma, which implies that the area was low during Palaeogene time. Scarce fission track data do not support the hypothesis of 2000 m of topography of the palaeo-EC during Palaeogene. If some early inversion of the former extensional basin occurred during Palaeogene, it would have generated a much lower local topography, probably limited to the borders of the former extensional basin without disturbing so much the sedimentary and palaeocurrents pattern in the LLA, MV and axial region of the palaeo-EC. A lower basin inversion topography of the former extensional basin alone is not enough to produce the observed Palaeogene subsidence. Thus additional components including topographic load of the Central Cordillera as well as remaining thermal subsidence after Mesozoic-rifting events can be proposed to explain the observed Palaeogene subsidence. 9.
MODELLING OF SUBSIDENCE PRODUCED BY THE COMBINED EFFECT OF (a) REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING, (b) FLEXURAL SUBSIDENCE PRODUCED BY THE TOPOGRAPHIC LOAD OF THE PALAEOCENTRAL CORDILLERA AND (c) FLEXURAL SUBSIDENCE PRODUCED BY LOCAL TOPOGRAPHY DUE TO PARTIAL INVERSION OF THE MESOZOIC EXTENSIONAL BASIN
As none of the hypotheses alone is adequate to explain the observed tectonic subsidence during Palaeogene times, I also tested whether a combination of all hypotheses to see if the combination of them could explain the observed tectonic subsidence. I modelled subsidence produced by the combined effect of: (a) remaining thermal subsidence after Mesozoic rifting, (b) flexural subsidence produced by topographic load of the palaeo-Central Cordillera and (c) flexural subsidence produced by local topography within the former extensional basin. To this aim I performed 2D and 3D modelling experiments.
9.1.
2D MODELS
Along 4 regional cross-sections (sections 2, 5, 8 and 12, see location i n Figure 3.2) I applied the following procedure: (a) From a series of stratigraphic columns or wells clo se to each section ( Figure 3.2), I calculated the observed tectonic subsidence using backstriping techniques, for six different time intervals from late Maastrichtian to Early Miocene; (b) I calculated also the remaining thermal subsidence after Mesozoic rifting events using forward modelling and all the parameters obtained from the study of Mesozoic rifti ng (Chapter 2); (c) Assuming local isostasy for each stratigraphic column or well and for each time interval, I calculated a subsurface load that would produce the same tectonic subsidence as the remaining thermal subsidence alone; (d) I modelled in 2D the flexural deflection of the lithosphere. For each regional cross-section and for each time interval, I modelled the effect of three different loads: (1) A subsurface load that would produce a subsidence effect equal to the remaining
109
Chapter 3
thermal subsidence; (2) A surface topographic load representing Palaeogene topography of the palaeoCentral Cordillera; and (3) Surface loads representing Palaeogene topography in some local areas of the former extensional basin. In this way I include in the flexural model the combined effects of remaining thermal subsidence after Mesozoic rifting, and flexural subsidence produced by loads in the palaeoCentral Cordillera and by local surface-uplifted regions related to basin inversion proce sses. I used the same values used in previous models for densities, crustal thickness, etc. (see Table 3.1 ). I systematically changed the magnitude of the topographic load of the palaeo-Central Cor dillera and both the location and magnitude of the topographic load of the former extensional basin as well as the values of effective elastic thickness until I reached the best fit between the observed and predicted Palaeogene subsidence. SECTION 5 FLEXURE
FLEXURE
) 4000. m ( 2000. n o i t 0. c e l f e -2000. d
4000.
) m ( 2000. n o i t 0. c e l f e d -2000.
-4000.
-4000.
-6000. 0.
500.
1000.
1500.
0.
EFFECTIVE ELASTIC THICKNESS ) m k ( t e e
500.
1000.
1500.
EFFECTIVE ELASTIC THICKNESS
80.
80. ) m k ( t e e
60. 40.
60. 40.
20.
20.
0.
0. 0.
500.
1000.
1500.
0.
500.
1000.
distance (km)
distance (km)
SECTION 3
SECTION 12
FLEXURE
1500.
FLEXURE
4000. ) m ( 2000. n o i t c 0. e l f e d -2000.
) m ( n o i t c e l f e d
2000. 0. -2000.
-4000. -4000. 0.
500.
1000.
1500.
0.
EFFECTIVE ELASTIC THICKNESS ) m k ( t e e
500.
1000.
1500.
EFFECTIVE ELASTIC THICKNESS
80.
) m k ( t e e
60. 40. 20.
80. 60. 40. 20.
0.
0. 0.
500.
1000.
1500.
0.
distance (km)
500.
1000.
1500.
distance (km)
Figure 3.29: 2D lithosphere flexural models assuming a positive topography restricted to the former extensional basin. The assumed topography fits best the observed thickness of Oligocene-Early Miocene deposits. Top panel: basement deflection and topography in meters. Dots represent th e observed thickness of Oligocene-Early Miocene deposits in meters. Bottom panel: effective elastic thickness in k m. Figure 3.2 shows the location of these sections.
110
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes 9.1.1. Results
The best results are presented in Figure 3.30 t o 3.33. Although the fit is not perfect I systematically changed each variable an d exhaustively tested many possibilities until I became convinced I had arrived to the best possible fit. The differences between the observed and predicted subsidences are short wave length effects that probably were produced by local relative rock-uplift or subsidence in fault-limited blocks. As such effects are not considered in the theoretical model, the model can obviously not predict them. The subsidence pattern of a partially inverted basin is characterized by short wave local subsidence variations produced by relative movement of faulted blocks. Table 3.2 shows the values and location of topographic loads as well as of the spatially variable effective elastic thickness necessary to produce the best fit between the observed and predicted subsidence. In order to approach the short wave length pattern of observed subsidence, it was necessary in many cases to apply two topographic loads close to the border faults limiting the former extensional basin. However, the number and location of loads is limited in the program. This pattern suggests that this topography could have been generated by early partial inversion of the border faults l imiting the former extensional basin. I also calculated the bending stress in the adopted thin elastic plate ( Figs 3.30 to 3.33). Bending stresses in all cases are within reasonable values under normal ro ck strength assumptions.
Sub-basin
Distance Maastrichtian from Early Paleocene Late Paleocene 60.9 Ma 54.8 Ma Romeral (km) EET * Topo EET Topo (km) (m) (km) (m)
Eocene 33.7 Ma
Oligocene Early Miocene 16.4 Ma
EET (km)
Topo (m)
EET (km)
Topo (m)
10 10 5 5 5 5 5 5 5 5 5 5 5 5 50 50 50 60
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
10 10 5 5 5 5 5 5 5 5 5 5 5 5 60 60 60 60
100 100 0 0 200
10 10 5
0 0 0
10 10 5
100 100 0
SECTION 2 Central Cordillera Middle Magdalena Valley " " Tablazo sub-basin " " " " " " Santander Massif " " " " Cocuy sub-basin " " " " Llanos foothills " " Llanos Orientales " " " "
0 270 271 350 351 355 380 416 417 424 437 438 509 523 524 555 556 800 0 257 258
1 1 1 1 1 1 5 5 5 5 5 5 5 1 5 50 50
100 100 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
10 10 1 1 1 1 5 5 5 5 5 5 5 5 5 60 60
100 100 0 0 100 100 100 100 100 100 0 0 0 0 0 0 0 0
1 1 5
0 0 0
10 10 5
100 100 0
0 0 0 0 0 0 0 200 200 200 0 0
SECTION 5 Central Cordillera Middle Magdalena Valley " "
ET: effective elastic thickness, Topo: palaeotopography *E Table 3.2: Effective elastic thickness and palaeotopography loads that best fit the observed subsidence during four different Palaeogene time intervals
111
Chapter 3
Sub-basin
Distance Maastrichtian from Early Paleocene Late Paleocene 60.9 Ma 54.8 Ma Romeral (km) EET * Topo EET Topo (km) (m) (km) (m)
Tablazo sub-basin " " Cocuy sub-basin " " " " " " " " " " " " Llanos foothills " " " " " " " " Llanos Orientales " " " "
300 301 367 368 402 405 417 445 446 452 453 460 480 481 485 486 800
Central Cordillera Middle Magdalena Valley " " Tablazo sub-basin " " " " Cundinamarca sub-basin " " " " " " " " " " " " " " " " " " " " " " Llanos Orientales " " " "
0 228 229 260 262 263 293 294 310 311 326 330 367 369 394 395 407 408 428 429 800 0 200 201 231 265 303 304
SECTION 8
5 5 1 1 1 1 1 1 1 1 1
Topo (m)
EET (km)
Topo (m)
5 5 5 5 1 1
5 5 5 5 1 1
5 5 5 5 5 5 60 60
0 100 100 0 0 0 0 0 500 500 500 500 500 500 500 0 0
50 50 50
5 5 5 5 5 5 50 50
5 50 50
200 200 200 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
10 10 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 50 50
100 100 0 0 0 200 200 200 200 0 0 0 0 200 200 0 0 0 0 0 0
10 10 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 50 50
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
10 10 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 50 50
5 5 5 5 5 5 5
100 100 0 0 0 0 0
10 10 5 5 3 3 3
100 100 0 0 0 0 300
10 10 5 5 5 5 5
0 0 0 0 0 0 0
10 10 5 1 1
5 5 5
0 100 100
EET (km)
50 50 50 50 0 0 0 0 0 0 0
10 10 1 1 1 1 1 1 1 1 1
5 5 5 5 1 1
Oligocene Early Miocene 16.4 Ma
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
10 50 50 50 50
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Eocene 33.7 Ma
100 100 0 0 0 0 0 0 0 0 0 0
1000 0 0 0 0
SECTION 12 Central Cordillera Southern Cundinamarca and Cocuy sub-basins " " " " " " " "
ET: effective elastic thickness, Topo: palaeotopography. *E Table 3.2: Continued
112
100 100 0 0 0 0 0
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
Sub-basin
Quetame Massif " " " " Llanos Orientales " " " " " " " "
Distance Maastrichtian from Early Paleocene Late Paleocene 54.8 Ma 60.9 Ma Romeral (km) EET * Topo EET Topo (km) (m) (km) (m) 318 319 345 350 351 354 360 800
5 5 5 5 50 50 50 50
0 0 0 0 0 0 0 0
3 3 3 3 3 3 50 50
0 0 0 0 0 0
Eocene 33.7 Ma
Oligocene Early Miocene 16.4 Ma
EET (km)
Topo (m)
EET (km)
Topo (m)
5 50 50 50 50 50 50 50
0 0 0 0 0 0 0 0
5 5 5 5 60 60 60 60
0 700 700 700 0 0 0 0
ET: effective elastic thickness, Topo: palaeotopography. *E Table 3.2: Continued Late Maastrichtian-early Paleocene:
( Fig. 3.30). To fit the total tectonic subsidence that occurred during this time interval, it was nec essary to apply topographic loads only in the area of the Central Cordillera. This result suggests that during this time the former extensional basin was not yet inverted, but it was still part of a foreland basin comprising the MV, palaeo-EC and LLA areas. The maximum amount of topography needed to fit observed d ata is 200 m, indicating a low topography, but it varies from north to south between 0 and 200 m (see Figures 3.30 and 3.3). Values for effective elastic thickness are very small (1 to 5 km) in the area of the former extensional basin, suggesting a weak lithosphere probably as a result of the Mesozoic rifting events. Also the inferred effective elastic thickness in the area of the Palaeo-Central Cordillera suggests a weak lithosphere. Only in the eastern LLA area values of effective elastic thickness of 50 km indicate the presence of strong lithosphere. These results can explain the fact that further deformation affected weak palaeo-Andean lithosphere but not the stronger LLA lithosphere. Horizontal stresses may also have had a significant role in the flexural history of the EC and on the lithosphere strength (Burov and Diament, 1995). However, in the actual situation, models do not require plate boundary forces or subsurface loads, and for simplicity I neglected their effect. Late Paleocene: ( Fig. 3.31) Results suggest low topography: up to 100 m in the area of the Central Cordillera an d also locally along the borders of the former extensional basin, with values up to 200 m. (Figures 3.31 an d 3.5). Early slight inversion of the border faults limiting the former extensional basin probably generated this topography, which did not significantly disturb the sedimentary and palaeocurrent pattern in the basin. The occurrence of local unconformities as reported by Sarmiento (1992, 1993) in the Guaduas Fm, might be related to this topography. Effective elastic thickness in the area of the Central Cordillera increased to 10 km, probably as a result of earlier deformation. The former extensional basin lithosphere remained weak (effective elastic thickness between 1 and 5 km), and prone to further deformation, whereas the eastern LLA lithosphere remained relatively strong (Effective elastic thickness of 50 km). Eocene: ( Fig 3.32) Models do not require topography at this time interval. If some surfaceuplift occurred, it was probably rapidly eroded (as indicated by evidences of erosion truncation of earlier structures in the Middle MV). The map of the early E ocene unconfor mity prepared by George et al. (1997), based on of geological maps and subsurface da ta (Fig. 3.6), shows: (a) a highly variable
113
Chapter 3
MAASTRICHTIAN EARLY PALEOCENE SECTION 2
SECTION 8
FLEXURE
500.
) m 100. ( n o i t c 0. e l f e d
FLEXURE
) m ( 0. n o i t c e -500. l f e d
-1000.
-100.
-1500. 0.
200.
400.
600.
800.
1000.
0.
EFFECTIVE ELASTIC THICKNESS 80. ) m k ( t e e
200.
600.
800.
1000.
EFFECTIVE ELASTIC THICKNESS 80. ) m k ( t e e
60. 40.
60. 40.
20.
20.
0.
0. 0.
200.
400.
600.
800.
1000.
0.
200.
distance (km)
400.
600.
800.
1000.
800.
1000.
800.
1000.
distance (km)
SECTION 5
SECTION 12
FLEXURE
FLEXURE
400.
400.
) m 200. ( n o i 0. t c e l f -200. e d
) m ( n o i t c e l f e d
-400.
200. 0. -200. -400. -600.
0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
80.
80. ) m k ( t e e
400.
) m k ( t e e
60. 40. 20.
200.
400.
600.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20.
0.
0. 0.
500.
1000.
0.
distance (km)
200.
400.
600.
distance (km)
Figure 3.30: Late Maastrichtian-early Paleocene 2D lithosphere flexural models assuming the combined effect of: (1) remaining thermal subsidence after Mesozoic rifting; (2) topographic load of the palaeo-Central Cordillera; and (3) local palaeotopography due to partial inversion of the Mesozoic extensional basin. Top panel: basement deflection and topography in meters. Dots represent the observe d thickness of late Maastrichtian–early Paleocene deposits in meters. Bottom panel: effective elastic thickness in km . Figure 3.2 shows location of these sections.
angular unconformity in the area of the Middle MV; (b) Truncated homoclines or broad relief folds in the area of the Upper MV and possibly (?) in the Santander Massif and borders of the original extensional basin; (c) Exhumated blocks with no deformation or very modest deformation in the former extensional basin; and (d) A par aconformity in the eastern LLA area. Values of effective elastic thickness required to fit the model s (Fig. 3.32), suggest the weakest lithosphere was located in the area of the former extensional basin (up to 5 km), with relatively weak lithosphere in the Central Cordillera area (10-km), and strong lithosphere in the eastern LLA area (50-km).
114
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes LATE PALEOCENE
SECTION 2
SECTION 8
FLEXURE
400.
FLEXURE
) 200. m ( n 0. o i t c e l f -200. e d
) m 200. ( n o i t 0. c e l f -200. e d
-400.
-400.
-600.
-600. 0.
80.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
200.
400.
600.
400.
600.
EFFECTIVE ELASTIC THICKNESS 80.
) m 60. k ( t e 40. e
) m k ( t e e
60. 40.
20.
20.
0.
0. 0.
500.
1000.
0.
200.
distance (km)
distance (km)
SECTION 5
SECTION 12
FLEXURE
FLEXURE 400.
200.
) m ( n 0. o i t c e l f -200. e d
) m 200. ( n o i t 0. c e l f -200. e d
-400.
-400.
-600. -600. 0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
500.
1000.
EFFECTIVE ELASTIC THICKNESS
80.
80.
) m 60. k ( t e 40. e
) m k ( t e e
60. 40.
20.
20.
0.
0. 0.
500.
1000.
0.
distance (km)
500.
1000.
distance (km)
Figure 3.31: Late Paleocene 2D lithosphere flexural models assuming the combined effect of: (1) remaining thermal subsidence after Mesozoic rifting; (2) topographic load of the palaeo-Central Cordillera; and (3) local palaeotopography due to partial inversion of the Mesozoic extensional basin. Top panel: basement deflection and topography in meters. Dots represent the observed thickness of late Paleocene deposits in meters. Bottom panel: effective elastic thickness in km. Figure 3.2 sh ows location of these sections.
( Fig. 3.33) in order to fit the data, the models require low topography in the area of the Central Cordillera (100 m) and the borders of the former extensional basin (Figs. 3.33 and 3.13). The required topography would be about 100 m in the western border of the basin, and 200 m in the eastern border of the basin. Locally in the Quetame Massif area 1000 m of topography would be necessary. If such topography existed, it was probably generated by inversion of the former extensional faults limiting the Mesozoic basin. However, such topography probably was not sufficiently continuous to significantly disturb the sedimentary and palaeocurrent patterns. Effective Oligocene - early Miocene:
115
Chapter 3
elastic thicknesses suggest weakest lithosphere (1 to 5 km), prone to further deformation was located in the area of the former extensional basin, with weak lithosphere in the Central Cordillera area (10 km) and strong lithosphere (50 to 60 km) in the eastern LLA area. EOCENE SECTION 2
SECTION 8
FLEXURE
FLEXURE 200. ) m ( 100. n o i t c 0. e l f e d -100.
) m 200. ( n o i t c 0. e l f e d
-200.
-200. 0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
500.
1000.
EFFECTIVE ELASTIC THICKNESS
80.
80.
) m 60. k ( t 40. e e
) m k ( t e e
60. 40.
20.
20.
0.
0. 0.
500.
1000.
0.
distance (km)
500.
1000.
distance (km)
SECTION 5
SECTION 12
FLEXURE
FLEXURE 150.
) m 200. ( n o i t c 0. e l f e d
) m 100. ( n 50. o i t c 0. e l f e -50. d
-200.
-100. -150. 0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
500.
1000.
EFFECTIVE ELASTIC THICKNESS
80.
80.
) m 60. k ( t 40. e e
) m k ( t e e
60. 40.
20.
20.
0.
0. 0.
500.
1000.
0.
distance (km)
500.
1000.
distance (km)
Figure 3.32: Eocene 2D lithosphere flexural models assuming the combined effect of: (1) remaining thermal subsidence after Mesozoic rifting; (2) topographic load of the palaeo-Central Cordillera; and (3) local palaeotopography due to partial inversion of the Mesozoic extensional basin. Top panel: basement deflection and topography in meters. Dots represent the observed thickness of Eocene deposits in meters. Bottom panel: effective elastic thickness in k m. Figure 3.2 sh ows location of these sections.
116
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes OLIGOCENE EARLY MIOCENE SECTION 2
SECTION 8
FLEXURE
FLEXURE
400. ) 1000. m ( n 0. o i t c e l f e -1000. d
) m 200. ( n o i 0. t c e l f e -200. d
-400.
-2000.
-600. 0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
EFFECTIVE ELASTIC THICKNESS ) m k ( t e e
60. 40.
60. 40.
20.
20.
0.
0. 0.
500.
0.
1000.
500.
distance (km)
distance (km)
SECTION 5
SECTION 12
FLEXURE
1000.
) m 500. ( n o i t 0. c e l f e d -500.
1000.
FLEXURE
) m 500. ( n o i t c 0. e l f e d
-500.
-1000.
-1000. 0.
500.
1000.
0.
EFFECTIVE ELASTIC THICKNESS
500.
1000.
EFFECTIVE ELASTIC THICKNESS
80. ) m k ( t e e
1000.
80.
80. ) m k ( t e e
500.
80. ) m k ( t e e
60. 40. 20.
60. 40. 20. 0.
0. 0.
500.
1000.
0.
distance (km)
500.
1000.
distance (km)
Figure 3.33: Oligocene-Early Miocene 2D lithosphere flexural models assuming the combined effect of: (1) remaining thermal subsidence after Mesozoic rifting; (2) topographic load of the palaeo-Central Cordillera; and (3) local palaeotopography due to partial inversion of the Mesozoic extensional basin. Top panel: basement deflection and topography in meters. Dots represent the observed thic kness of Oligocene-Early Miocene deposits in meters. Bottom panel: effective elastic thickness in km. Figure 3.2 shows location of these sections.
9.2.
3D MODEL
Using the results of the 2D models, I carried out a 3D model experiment for the Oligocene– Early Miocene time interval applying topographic loads in the palaeo-Central Cordillera and the borders of the original extensional basin, as suggested by the 2D results, to calculate the flexural deflection of the lithosphere. Figure 3.8b shows the calculate d deflection. Comparison of this result (Figure 3.34) with the observed subsidence patter n (Fig. 3.13) shows a better fit than the result obtained
117
Chapter 3
in a similar 3D model without applying topography loads to the borders of the former extensional basin (Fig. 3.28). Serrania de Perija
1400
Maracaibo Basin
e s d A n i d a r M e
Cucuta s a c u L n a S e d i a n a r r e S
1300
Medellin
1200
Bucaramanga Arauca
a le r i o rd C
100 0
Manizales
Ibague
Llanos Orientales Basin (LLA)
Tunja
1100
1000
Barinas Basin
Yopal
Bogota
-
0
100
200 Km
LITHOSPHERE FLEXURE PRODUCED BY A UNIFORM LOAD ON THE CENTRAL CORDILLERE, TABLAZO AND COCUY (FOOTHILLS) INVERTED SUB-BASINS (m)
Villavicencio 300
900
l t ra e n C
200
Neiva
800
800
900
a e n e d r a a i n c a a M r r e a S L
1000
100 San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 3.34: 3D lithosphere flexural models. 3D flexural defection of the lithosphere produced by a combined load represented by topography in the palaeo-Central Cordillera and also in the inverted Tablazo-Magdalena and Cocuy sub-basins of the former extensional domain. See these results with thickness maps of Oligocene-Early Miocene deposits for a comparison (Fig. 3.13).
As a conclusion, this result confirms the 2D models, also suggesting development of some local topography along the borders of the former extensional basin, probably as a result of inversion of the original extensional faults limiting the basin. If some extensional faults were reactivated it was probably because their strength was lower than normal rock strength. Calculated bending stresses assuming a thin elastic plate are lower than normal rock strength values. Such topography was low and did not significantly disturb the sedimentary and palaeocurrent pattern in the Palaeogene basin. Modelling results require very low values of effective elastic thickness in the area of the former extensional basin. The Mesozoic rifting events reduced significantly the strength of the lithosphere, making it very prone to Palaeogene deformation and to further Andean deformation. Low values of effective elastic thickness are usually associated with fault-controlled upper crustal flexure patterns (Van Wees and Cloetingh, 1994). These authors have shown that superposition of low EET values with high EET flexure effects, supports a multilayered rheological control on continental rifting. Probably similar effects also are applicable to the early basin inversion processes in the palaeo-EC of Colombia.
118
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
The differences between the observed and predicted subsidence are short-wavelength effects probably produced by local relative surface-uplift of fault limited blocks.
10. COMPARISON OF MODEL RESULTS WITH FISSION TRACK DATA AND OTHER EVIDENCE OF UPLIFT AND DEFORMATION Flexural models suggest local palaeotopography (surface-uplift), while erosion-related unconformities and petrographic evidence of erosion of some areas indicate local exhumation. If During Palaeogene time in the study area, local uplift of the earth’s surface occurred and also denudation/exhumation occurred, the amount of rock-uplift must be greater than the surface-uplift estimated from flexural models (see equation 3.1, England and Molnar, 1990). Are the model results supported by additional evidence of exhumation? To answer this question I summarise from literature fission track data and other evidence of exhumation and deformation. Finally I compare model results with these data.
10.1. FISSION TRACK DATA AND OTHER EVIDENCE OF EXHUMATION AND DEFORMATION Although fission track data represent cooling and denudation, they can be related to denudation/exhumation in combination wit h knowledge of the geothermal gradient ( e.g. Andriessen, 1995; Rohrman, 1995). Figure 3.35 shows the distribution of fission track data and deformation/exhumation evidence in northwestern South America, compiled from literature. Numbers illustrate the localities on Figure 3.35. In most of the mentioned cases fission track data must be interpreted with great care because track lengths have not been measured ( e.g.A ndriessen, 1995). 1. Mérida Andes (Venezuela) . Kohn et al. (1984) studied apatite, zircon and sphene fission tracks from 45 samples of the Merida Andes in Venezuela. Zircon ages range from 60 Ma to 172 Ma with a strong group in the range of 81 to 113 Ma. These authors interpreted zircon ages older than 61 Ma as mixed ages. Two sphene ages at 139 and 159 Ma also were interpreted as mixed. However, Van der Wiel (1991), Van der Wiel and Van der Bergh (1992a,b) and Van der Wiel and Andriessen (1991) reinterpreted some of the 81 to 113 Ma ages as related to a tectonic event that produced a small amount of exhumation. Zircon ages of 60 and 61 Ma are interpreted as exhumation ages suggesting initial rock-uplift during early Paleocene. According to these authors this rock-uplift was sufficient to make the Mérida Andes positive but not enough to bring the basement rocks through the closure temperature for apatite (Kohn et al., 1984). However this is against the petroleum occurrences in the Barinas Basin, which are assumed to derive from the west of the Mérida Andes (Roure, personal communication). Apatite ages range from 24 Ma to 1.4 Ma. Kohn et al. (1984) interpreted t hem as cooling ages related to exhumation and erosion of cover rocks. Based on the apatite ages, the y also suggested that exhumation of the Mérida Andes started in the leading (northwestern, location 1a s ee Fig. 3.35) mar gin during the Oligocene to Mioc ene, followed by exhumation of the trailing (southeas tern, location 1b see Fig. 3.35) margin in the Late Miocene, and later by rapid exhumation of the central axial Andes (location 1c see Fig. 3.35) during Pliocene-Pleistocene time. A plot of apatite ages versus sample elevation approximates a straight line suggesting an exhumation of approximately 2300-m during Pliocene. These authors assume updoming of isotherms, because the estimated exhumation rate of 800 m/Ma exceeds accepted rates of thermal diffusion in rocks. Also they suggested that the time interval of exhumation is likely to be greater than indicated by the range of ages. 2. Sierra de Perijá . Shagam et al. (1984) studied apatite, zircon and sphene fission tracks from 9 samples from this area. Zircon ages range from 69 to 127 Ma. One sample gives the crystallisation age of felsic volcanics (120- 122 Ma), whereas 2 or 3 samples are interpreted as indicative of exhumation at the end of Cretaceous-Paleocene time and the remaining are interpreted as mixed ages. Apatite ages range between 2.7 Ma and 27.2 Ma being interpreted as exhumation during late Oligocene (27-22 Ma) in the southeast piedmont (location 2a see Fig. 3.35) followed by exhumation during early to middle Miocene (19-14 Ma) in the western piedmon t (location 2b see Fig. 3.35) and Middle Pliocene
119
Chapter 3
(3 Ma) in the whole Sierra de Perijá (Shagam et al., 1984). One sphene age of 113 Ma was interpreted as a mixed age. Toas Islands Zircon: 120 to 93 Ma. Apatite: 13 to 12 Ma Shagam et al. (1984)
Caribbean Sea
Sierra Nevada de Santa Marta
Sierra de Perija
P a n a m a
n a e c O c i f i c a P
2b 2c 2a Maracaibo Basin Mer ida
Andes
ra i l le rd o C rn s te e W
l n t ra e C
) V 3a M ( y e 3b l a l l le r V a ) C rdi n a E ( 3c e o a C d l r a l e g l 8 i a d r M o 4 6 n C 5 r t e s a E
9b 9c
7
1a 1c 1b
Caribbean Mountains Zircon: 49+6, 42+5.3, 19.7+2.1 Ma. Apatite: 6.1+1.3 M/a. Kohn et al. (1984) Barinas Basin
Venezuela
) L A L ( s e l t a n e i r O s o l a n L
9a
Guyana Shield
Putumayo Basin
0
100
200 km
Ecuador
Napo and Cutucu Mountains Cooling and uplift events: 40 and 12 Ma. Marksteiner and Aleman (1997)
Brazil
Cordillera Real and Occidental Zircon ages: Cretaceous to Miocene Apatite ages: Paleocene to Miocen e Steinmann et al (1996)
Figure 3.35: Location of fission track data and other evidence of exhumation. Numbers refer to location discussed in the text. 3. Santander Massif. Shagam et al. (1984) also studied apatite and zircon fission tracks from 12
samples of this massif. Zircon ages range between 61 and 109 Ma and were interpreted to indicate exhumation of the massif during the end of Cretaceous-Paleocene (with the older ages being mixed). Apatite ages range between 18.9 and 3.8 Ma, indicative for exhumation during middle Miocene (19-14 Ma) in the central part of the massif (location 3b Fig. 3.35) and late Miocene to early Pliocene (7-4 Ma) in the central (location 3b Fig. 3.35) and northern (location 3a Fig. 3.35) parts of the massif. Toro (1990) reported an apatite fission track age of 30.8±5.8 Ma from a granite sample taken along the
120
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
Soatá-Onzaga road in the southernmost Santander Massif (location 3c Fig. 3.35). Crowllie, 1989 (in Toro, 1990), who performed the analysis, reported that the confined track lengths are shorted than expected and interpreted the age as a mixed one. Forward modelling of the measured ages and lengths according to the geological history suggest burial heating between 180 and 30 Ma followed by exhumation and cooling between 30 Ma and present time (Crowllie, 1989 in Toro, 1990). 4. Floresta Massif. Toro (1990) reported an apatite fission track age of 22.3±4 Ma from the Belén granite. Crowllie, 1989 (in Toro, 1990), who performed the analysis, reported that the confined track lengths are shorted than expected and interpreted the age as a mixed one. Forward modelling of the measured ages and lengths according to the geological history suggest burial heating between 180 and 30 Ma followed by exhumation and cooling between 30 Ma and present (Crowllie, 1989 in Toro, 1990). 5. Eastern flank of the Eastern Cordillera. According to Hossack et al.( 1999), based on apatite fission track data, the Chámeza thrust, which is the eastern margin of the EC, began its initial exhumation at 25 Ma. However, the main deformation in the eastern foothills did not begin until 15 Ma, continued to 3 Ma, and may still be active at present. According to these authors the only obvious synorogenic sediments that were deposited in the eastern foothills around growing thrust structures are the youngest post 3 Ma Pliocene-Pleistocene deposits. Apatite fission track data for the more internal parts are younger and define a general piggy-back development of the foothills from 15 Ma to 3 Ma (Hossack et al., 1999). In the LLA foothills Rathke and Coral (1997) show ed evidence of Oligocene early development of the Cupiagua oil field structure. C orredor (1997) has re ported similar evidence of structural deformation in the eastern flank of the E C. Figures 2.23 a nd 3.14 s how that incipient inversion of Cretaceous normal faults occurred dur ing Oligocene time. Confi dential data from Petroleum Companies suggest that since late Miocene (10-12 Ma) and during the Pliocene, massive regional denudation occurred. Probably rock-uplift, exhumation and deformation started in the marginal thrust fault systems and progressively moved toward the inner axial zone of the Cordillera. ranquet et al.( 1996,1999a,b) and Branquet (1999) 6. Western flank of the Eastern Cordillera.B have shown that on the western flank of the Cordillera emerald deposits are genetically linked with tear faults and associated thrusts developed during a compressive phase that occurred at the EoceneOligocene time boundary (38-32 Ma). Such an age is based on 40Ar/ 39Ar and K/Ar dating of syngenetic green muscovite crystallized on emerald-bearing vein wallrocks (Cheilletz et al., 1993, 1997). However, Romero et al. (2000) based on Rb/Sr analysis questioned the 38-32 Ma age and proposed an age close to the Cretaceous-Paleocene boundary (67 Ma for the western emerald belt and 61 Ma for the eastern emerald belt) According to Cheilletz et al., (1993, 1997) emerald precipitation occurred at temperatures of 300 °C at a burial depth comprised between 4000 and 5000 m (Branquet, et al., 1999a). Emeralds occur within Valanginian limestone and Hauterivian black shale (Branquet, 1999). If the stratigraphic thickness in this region between the Valanginian-Hauterivian boundary and the top of the Cretaceous is approximately 7000 m, as reported by Rubiano (1989) then between 2000 and 3000 m of Cretaceous sediments would had been eroded during Eocene/Oligocene time, if Cheilletz et al. age is accepted, or close to the Cretaceous-Paleocene boundary if Romero’s age is accepted. In the Middle MV and western EC foothills “AFTA, vitrinite reflectance data and eastward thinning of Oligocene units (Mugrosa and Colorado Fms) constrain the age of the first uplift event of the foothills region to the time between 34 to 30 Ma. At this time Cretaceous rocks began to cool from a maximum temperature of 180 º. Younger Upper Oligocene-Lower Miocene rocks preserve a syntectonic stratal record of instantaneous limb rotation, associated with layer parallel shearing. Several oil pr oducing anticlines e.g. Provincia and Lisama belong to this generation of folds” (Gomez et al., 1999, Fig. 3.15). “Miocene sediments were originally deposited across the foothills area of the present EC as indicated by AFTA calculations of the eroded sedimentary column and balanced structural sections” (op cit ). “Diachroneity of westward breakthrough of the EC is indicated by the age of a second thermal event. Cooling from maximum temperatures in the range of 110 to 120° C started between 15 and 5 Ma in the southern part of the basin and between 5 to 0 Ma in the northern Middle MV according to AFTA and vitrinite reflectance analyses” ( op cit ). “The main uplift of this mountain range and complete conversion of the Middle MV into an intermontane basin occurred after 6.2 Ma.” (op cit ).
121
Chapter 3
Based on outcrop, chronological data from volcanic ashes and palaeontology, fission track and subsurface data Gomez et al. (1999) interpreted that during Eocene-Early Miocene communication between the Middle MV and LLA basin to the east was partially interrupted by low discontinuous hills, which resulted from incipient inversion of Mesozoic grabens in the area of the EC. Growth stratal relations and time transgressive unconformities document this ( op cit ). “In the Southern Middle MV progressive forelimb rotation of west-verging anticlines, indicate that inclined shear was the likely mechanism of Palaeogene deformation of the EC. Northward directed palaeocurrents attest to the Eocene-Oligocene confinement of the MV between the Central Cordillera and EC. Upper PlioceneLower Miocene strata finally overlapped the EC uplift as the sedimentation sourced by the Central Cordillera overcame uplift rates of the eastern folds. Eocene-Lower Miocene basin fill also onlapped over the Paleocene alluvial fans and the pediment surface that resulted from erosional retreat of the Central Cordillera front” ( op cit ). For the northern Middle MV Gomez et al. (1999) interpreted that “Paleocene and older r ocks were highly deformed during Eocene as the central Cordillera front propagated into the basin ( Fig. 3.7). Net westward retreat of the Central Cordillera front has induced an eastward-dipping unconformity, which is the base on the onlapping Eocene to Neogene sediments. Chronology and kinematics of EC deformation events in the northern MV are not the same as in the south. Palaeocurrent and provenance analyses indicate that central areas of the EC were uplifted during sedimentation of the lower part of the Eocene La Paz Fm” ( op cit ). An event of westward thrust faulting of probable Eocene age has been identified in the central part of the Middle MV by RestrepoPace et al. (1999a,b). “Westward migration of EC deformation continued during Oligocene-Early Miocene, as a series of fault propagation faults” (Gomez et al., 1999). 7. Sabana de Bogotá (axial zone of the Eastern Cordillera). Palynological and fission track data on zircons obtained from ash layers indicate that the fluvial-lacustrine sediment record of the Sabana de Bogotá area registers major tectonic surface-uplift for the period between 5 and 3 Ma (Andriessen et al., 1993). This will be discussed in detail in Chapter 4. 8. Central Cordillera. Jaramillo (1978, 1981) reported fission track ages of 10.5±1 Ma (Late Miocene) on apatite and 58±6 and 62.4±3.6 Ma (Paleocene) on zircon, from the Manizales Pluton at Fresno in this area. Based on fission track data and outcrop studies Gomez et al. (1999) proposed that the “Central Cordillera formed by transpressional strike-slip deformation beginning in the Campanian. Fission track cooling ages of the Mariquita stock, a granitoid pluton in the eastern flank of the mountain range, indicate slow cooling at rates of 2.8 º C/Ma between the Campanian (77.6±7 Ma, zircon) and the Oligocene (32.0±6.2 Ma, apatite), and 2.5 º C/Ma since the Oligocene” ( op cit ). According to these authors during Paleocene eastward migrating deformation of the Central Cordillera tilted the mountain front alluvial fans (Hoyón Fm). In the northern part of the Central Cordillera Toro et al.( 1999) reported fission track ages from 55 samples. 19 samples from the metamorphic and igneous (Antioquia Batholith) basement give zircon ages between 75 and 35 Ma, while 16 samples correspond to Paleocene. These apparent ages indicate that the basement cooled below 300 ºC during Paleocene. According to these authors an older age obtained of 185±5.5 Ma could indicate that the entire basement could not have been thermally affected above 300 ºC during all the Cretaceous thermal events. Ages from volcanic zircons are comprised between 0.35±0.05 Ma and 6.19±0.23 Ma with a peak between 1 and 4 Ma. Maastrichtian unroofing of the Central Cordillera sourced quartzite and chert pebbles for the Cimarrona Fm (Gómez and Pedraza, 1994). According to Campbell (1974) and Anderson (1970, 1972) conglomeratic Cenozoic deposits in the MV recorded exhumation pulses of the Central Cordillera during Eocene and Oligocene. Vaning exhumation of the Cordillera during early to middle Miocene induced widespread muddy sedimentation (La Cira Colorado Fm) that recorded the beginning of explosive volcanism in the magmatic arc of the Central Cordillera (Van Houten, 1976). Episodes of active exhumation and volcanism of the Central Cordillera were recorded during 13.5 to 11.5 Ma (Honda Gp), and approximately 1.4 Ma (Neiva Fm) in the Upper MV (La Venta area, Guerrero, 1993). From the northern part of the Central Cordillera near Medellín, extensive remnants of at least three uplifted planation surfaces have been recognized by Page and James (1981), the Pre-Central Cordillera Erosion surface (Pre-S-1) over 3000 m, the Cordillera Central erosion surface (S-1) around 3000-2500 m, and the Río Negro surface (S-II) around 2200 m. The tilted S-1 surface is found to underlie the Miocene Honda Fm. By combining limited palaeomagnetic evidence on terrace deposits
122
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
and dissection depth of major rivers, Page and James (1981) calculated ages 22-18 Ma for the development of the S-1 surface, which is in agreement with the 14-16 Ma age found for volcaniclastic components in the Honda Fm (Setoguchi and Rosemberg, 1985). The Río Negro surface is probably older than 3 Ma. According to Soeters (1981a), the volcaniclastic Mesa Gp was dated palinologically in the Pliocene (Dueñas and Castro, 1981) and Thouret et al.( 1985) obtained 4.3±0.4 and 3.5±0.4 Ma KAr ages from it. The Río Negro surface was truncated therefore around 4 Ma. 9. Garzón Massif. Van der Wiel (1991) and Van der Wiel and Andriessen (1991) studied apatite, zircon and sphene fission tracks from 14 samples of the Precambrian metamorphic and Jurassic igneous rocks of the massif. Samples are from San Vicente del Caguán (11a), Tres Esquinas Ventanas road (11b) and Guadalupe-Florencia road (11c). Zircon ages range between 715 and 89 Ma and sphene ages range between 183 and 146 Ma. Apatite ages range between 13.9 and 9.2 Ma. Using several geological constraints and thermal fission track modelling to differentiate between mixed ages and cooling ages she interpreted three exhumation and cooling events. The first exhumation event of approximately 10.000m occurred 900 Ma ago probably related to orogenic processes. Between 850 and 210 Ma the Garzón Massif was denuded and eroded. The second differential 3500 m exhum ation event of faulted blocks occurred during the Cretaceous around 100 Ma. I discussed these ages in Chapter 2. (Kohn et al., 1984 and Shagam et al., 1984 interpreted those ages as mixed ages). Finally between 12 Ma and present time the massif was exhumated by approximately 6500 m. After the first N eogene (12 Ma) exhumation pulse a second pulse occurred 6.4 Ma ago. The rate of very fast exhumation must have exceeded the rate of thermal diffusion, producing thermal updoming of the isotherms, and the ages should be interpreted as cooling following exhumation (Van der Wiel, 1991). Westward palaeocurrent directions in the Villavieja Fm of the Upper MV indicate that the southern EC was already a mountain range that completely closed the basin in its eastern side at about 11.8 Ma (ages from 40Ar/39Ar dates and magnetostratigraphy, Guerrero, 1993). Changes of these palaeocurrents indicate that exhumation of the southern EC began at 12.9 Ma (Guerrero, 1993). 10. Upper Magdalena Valley. Schwabe et al. (2001) reported zircon fission-track ages from the Saldaña Fm. in this area. According to these authors Palaeozoic ages (235 to 395 Ma) represent reworked zircon grains derived from Precambrian rocks of the Garzón Massif. Jurassic ages (between 136 and 176 Ma) are in agreement with the age of the volcanic rocks of the Saldaña Fm. Cretaceous ages (100 ± 4 Ma) are in agreement with the age of the Yaví Fm. The presence of reworked Jurassic zircons within the Neogene Honda Fm. indicates erosion of uplifted Jurassic rocks as a local source area during the Neogene. In conclusion the fission track ages combined with geological evidence suggest the following: (1) Fission track ages and other geologic evidence indicates initiation of Andean exhumation of blocks at the end of Cretaceous-Paleocene with local phases of exhumation following during Eocene to Miocene time, terminating with regional exhumation in all blocks in the PliocenePleistocene. Shagam et al.( 1984) interpreted this as the result of uncoupled crustal blocks being uplifted in response to local stress until regional compression led to interlocking of the blocks and their simultaneous rock-uplift during the Pliocene-Pleistocene time. (2) Geological evidence suggests that Palaeogene exhumation of the Central Cordillera was regional, supplying detritus to the eastern area. In the area of the Mesozoic extensional basins exhumation was only local and probably of low elevation as suggested by local evidence of detrital source areas. (3) In the Perijá and Mérida Andes fission track data suggest initial Oligocene exhumation of one of the flanks, later Miocene exhumation of the other flank and finally exhumation of the whole area including the axial central region during Pliocene time. Data suggest also a similar picture for the EC. Deformation, and probably rock-uplift of the western flank started during Eocene/Oligocene (38-32 Ma) and possibly Paleocene. Rock-uplift and deformation of the eastern flank started during Oligocene and followed during Miocene (25 to 10 Ma) and regional rock and surfaceuplift including the central axial Sabana de Bogotá area during the Pliocene (5-3 Ma).
123
Chapter 3
10.2. COMPARISON WITH MODEL RESULTS Both model results and fission track data suggest a rock-uplift history starting at the end of Cretaceous/Paleocene and following during Palaeogene time. However rock-uplift was moderate and only affected tectonic blocks. Particularly model results and fission track data suggest initial moderate rock-uplift in the flanks of the EC as expected in a scenario of inversion of the former extensional fault systems delimiting the Mesozoic basin. As a first approach in a regional view, therefore, fission track data seem to support model results. However, if a very moderate surface-uplift of 100 to 200-m occurred in the flanks of the palaeo-EC as suggested by the models during Paleocene, it was not recorded by fission track data. A possible explanation for this is that such a low topography was not enough to produce enough exhumation and cooling. Because the lowest temperature associated with a cooling age is 50°-100° C (Faure, 1986), the amount of exhumation determined from cooling ages is likely to be at least 1-3 km (England and Molnar, 1990). Fission track data only suggest Paleocene exhumation of the Santander Massif and there is evidence that at that time the massif started to erode and supply detrital sediments (Fabre, 1986, 1987) Deformation/uplift history during Oligocene and early Miocene seems to agree with model results. However, more data are necessary to test model results and define in detail the Palaeogene uplift history of the EC.
11. STRUCTURAL KINEMATIC MODEL OF PALAEOGENE DEFORMATION, INCLUDING SEDIMENTATION AND EROSION EVENTS In the flexural models we have recognized some local differences between the observed and predicted subsidence. We interpreted these short wave length effects as produced by local relative surface-uplift of fault limited blocks. Basin-wide scale flexural models can not predict them. As an additional modelling test to try to understand these short wave effects of the Palaeogene subsidence uplift pattern of the EC and to test if incipient inversion of the Mesozoic extensional basin is a viable process consistent with available stratigraphic and structural data, I modelled the kinematic structural deformation, sedimentation and erosion during Palaeogene along the regional cross-section of the EC published by Cooper et al. (1995). This model incorporates structural geology to try to bridge basin wide and sub-basin local scales. I applied a 2D structural kinematic model of deformation, assuming inversion along the border faults of the former extensional basin, including also sedimentation and erosion events. Sedimentation, erosion and active compressional deformation processes are obviously interrelated. The purpose of such modelling is to reconstruct step by step the structural, sedimentary and erosion evolution of the basin during Palaeogene time. These models can be useful to infer the deformational history based on the knowledge of the sedimentary record. In this model I accepted the hypothesis of early contractional deformation of the former extensional basin during Palaeogene time, as suggested by local evidence (Restrepo-Pace, 1999a,b; Gomez et al., 1999) and previous model results. I estimated the amount of shortening necessary to generate local low topography as predicted by flexural models and also local subsidence to create accommodation space for the observed Palaeogene sedimentary record.
11.1 STRUCTURAL, SEDIMENTARY AND EROSION MODEL I used the program Thrustpack version 6.2 developed by Sassi et al. (1998) to model the structural history of the EC during Palaeogene time. I applied the model to a regional balanced crosssection. The program reconstructs through time the kinematic, structural, sedimentary and erosion history along a cross-section. Such a reconstruction requires knowledge of the initial geometry of the system before tectonic deformation occurred and the final deformed geometry (Sassi et al., 1998). The initial and final geometries are usually taken from a previous balanced cross-section interpretation and its restored state before deformation. The program simulates the kinematics of deformation, sedimentation and erosion in a forward sense, starting from the initial undeformed section and ending to the final, present day deformed section (Sassi et al., 1998). Because a good correspondence between the two cross-sections can strongly depend on the fault geometry assigned in the present day deformed
124
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
cross-section, it is necessary to use a section where the fault geometry at depth has been interpreted. I selected the regional balanced cross-section interpretation of Cooper et al. (1995) because it is the only section that includes an interpretation of faults at depth until reaching a deep detachment in the upper boundary of the lower crust. The program Thrustpack version 6.2 allows the kinematic structural modelling of the geometry of structures using both the fault-bend-fold deformation mechanism with kink fold geometry (Suppe, 1983) and the Foldis algorithm (Divies, 1997). In the Foldis algorithm the progressive folding of rock layers is performed using a model that allows discrete slip between layers and strain concentration in fold hinges. It simulates flexural slip mechanism that produces curved fold geometry and it can include compaction of the lithologies due to an increase of burial depth trough time (Sassi et al., 1998). A complete description of the algorithm Foldis can be found in Divies (1997). The sedimentation and erosion simulation also is purely geometric and consists in the addition or erosion or material up to a new earth surface defined by the user as a poly-line on the cross-section. The program also allows simulation of thermal conductivity, organic matter maturity, generation and expulsion of hydrocarbons from an oil source rock. However, in this study I only used the structural, sedimentation and erosion module, which requires as input data the initial undeformed cross-section geometry, its dimensions, the definition of each stratigraphic layer (including its properties), the initial topographic surface and the fault trajectories geometry. All layers and faults are defined as poly-lines. Faults should be defined starting from the most internal (hinterland) toward the most external (foreland). Each fault block or thrust sheet is defined, together with certain amount of displacement per time increment defined by the user, and applied to each fault block or thrust sheet starting with the autochthonous block. For each time increment the user can apply displacement to each fault block as well as sedimentation and erosion events. All displacements are referenced to the autochthonous block. Large-scale subsidence and uplift such as flexural effects are introduced as a file indicating the rock-uplift or subsidence as a function of the horizontal distance along the section. A complete description of the program can be found in Sassi et al.( 1998). In order to apply the program to the regional balanced cross-section by Cooper et al. (1995) I divided the section in tw o parts. I defined the faults and thrust sheets in each half of the section as shown in Figure 3.36. I selected similar time intervals as those applied in the flexural models. For each fault block (thrust sheet) and for each time interval, I applied several combinations of horizontal contraction displacements until I found acceptable displacements that can be compared to the observed subsidence or modelled palaeotopography. The amount of shortening in kilometers applied to each thrust sheet is shown in Table 3.3. I tried to generate an amount of topography similar to that suggested by flexural modelling and an amo unt of subsidence similar to that of the observed Palaeogene sedimentary record.
11.2. RESULTS After applying several combinations of contractional displacements to each thrust sheet for each time interval to best fit the obser ved subsidence and the calculated topography from flexural models, I arrived to the results shown in Figure 3.37. In this Figure the two halves of the section have been assembled in a single section, some minor mismatch between the two parts have been artificially eliminated in the drawing. This mismatch was produced by dividing artificially the original section in two parts and applying eastward displacement to the eastern part relative to autochthonous block in the eastern LLA, and westward displacement to the western part, relative to a western autochthonous in the western Middle MV ( Figure 3.37). In nature the less deformed autochthonous block is the eastern LLA, which is part of the South American craton, and all the Palaeogene deformation occurred by relative displacement toward this more stable block. The wes tern Middle MV probably was the most deformed block during Palaeogene, as evidenced in subsurface ( Fig. 3.15), contrary to this artificial assumption. However, the division of the section in two parts o nly was a necessary artifact to handle the section with the program. A simple assumptio n was applied reducing the number of active faults and their geometry, as shown in Figure 3.36.
125
Chapter 3
E s o n a l L
1
2
) k c o l b s u o n o h t h c o t u a e v i t a l e R ( 0
n o t i n t e c m e e s c e a h l t p s f o i f d l l a a h n o n i r s t s e s e r a p e m e o h c t n d r i a s w k t c s l o a b E f o
m k 0 0 1
4 3
n i s a b l a n o i s n e t x e c i o z o s e M
5
6
7
0 8
9
W
y e l a V a n e l a d g a M e l d d i M
0 1
) k c o l b s u o n o h t h c o t u a e v i t a l e R ( 1 1
n o l n i a i t n s c o k e s i s c e s l o h e r b t p f f o o f m l o t n a c e h d r m n r a e e w c t t a s s l p e e i s w W d e h t
126
s n d t e e r r e a o b p t s o s e a r t w h a r o k e t c l l n l o i i d b d r o e i c C d i v n n i o t r d c e t e n t s a e e E b r o e s t h a t h e e f h o n s o t n i t o c s i u t e r c s h e t s d e h s c s m a o r e r o c f f l e o a d t n o n u n i g e e e h r T m e c e . a ) h l t 5 p s o 9 i t 9 d 1 g , . A n l i l l a . e t d d e . t o e g a r n n i m e i l p l k m e o c o i l d a p C e o t s e m r m u e g r o r w i n h f T ( s r u r t l d o s u u f o a ” e d f k c e c i a e l o l t p e b r m p o a C S s s f . u o n o l y n o t i d o t n n h p e e t h m e d c n u h o s t e t s p t a a e u a “ d e d n t e e i a i h f t i t s d l e o p d l l e e t m e i m d v S r o i t : o a m f 6 e e l 3 e . d r 3 n r e d u e e w i r s t l u t i g a p i o h F t t p a
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
11
Western Part 10 9 8 7
Sh
6
5
4
0
0
0
0
0
0
0
0
0
0
0
End Paleocene 54.8 Ma
0
0
-3
-3
-3
-3
1
1
1
1
End Eocene 33.7 Ma
0
0
2
2
2
0
0
3
3
3
Thrust Sheet End Cretaceous 65 Ma
End Early Miocene 16.4 Ma Sh: shortening
-3.8 -3.8 -3.8 -3.8 -4
-6
-6.3 -6.3
Eastern Part 3 2 1
0
Sh
Total Sh
0
0
0
0
0
0
0
4
4
2
0
0
0
2
6
3
0
0
0
3
9.3
Table 3.3. Amount of shortening in kilometres applied to each thrust sheet in the Thrustpack program. Positive numbers indicate left to right displacement, negative values indicate opposite sense of displacement.
Particularly the deep detachment interpreted by Cooper et al.( 1995) below the MV was not considered in the model. Following Cooper et al. (1995) interpretation I applied displacement only to those faults interpreted by these authors as active during compression tectonics. For example I applied displacement to the Chámeza Fault considered to be active during Tertiary compression, but I did not applied displacement to the Guaicáramo Fault interpreted as a Mesozoic extensional fault passively transported with a short-cut basement block during Palaeogene compression deformation. A consequence of this interpretation and the geometry of faults at depth is the absence of a clear inversion of the faults defining the eastern border of the former extensional basin, contrary to the results suggested by flexural models, which are more in agreement with the structural cross-section interpretation by Colletta et al. (1990) who interpreted inversion of the Guaicáramo Fault. The results show that more clear inversion structures were developed in the western border of the former extensional basin. Although some Mesozoic normal faults were passively transported with short-cut basement blocks ( e.g. Las Esmeraldas Fault, ESRI and Ecopetrol, 1994), flexural models support their inversion (as interpreted by Colletta et al., 1990), which is more probable in a scenario of transpression. Normal faults that usually develop at dip angles close to 60° are not inverted in frontal compression, but they are inverted in oblique compression. Because of the very reduced thickness of the Palaeogene deposits compared to the deep detachment interpreted for the section it is difficult to reproduce with accuracy the geometry of these Palaeogene deposits, which at least partially have been exaggerated in the model. Similarly the topography in the model has been at least partially exaggerated compared with that suggested by flexural models, even when applying considerable amounts of erosion. An important result of the model is the relatively small amount of total shortening (up to 9.3 km at the end of early Miocene) which has produced local topography and subsidence greater than those observed or estimated through flexural models. Therefore, assuming the dip angle of the faults is as interpreted by Cooper et al. (1995), the estimate of 9.3 km of shortening at the end of early Miocene is an upper limit. Obviously the amount of shortening necessary to generate a fixed amount of rock-uplift/subsidence increases if the dip angle of the faults decreases.
127
Chapter 3
End of Cretaceous (65 Ma)
End of Paleocene (54.8 Ma) 4 km End of Eocene (33.7 Ma) 6 km End of Early Miocene (16.4 Ma) 9.3 km Present (0 Ma)
La Sal in a Fault
B o y a c a F P e s c a F Chameza F Guaicaramo F CusianaF
70 km
100 km Figure 3.37: Kinematic structural models obtained with the Thrustpack program, simulating the evolution of the Eastern Cordillera during Palaeogene time. The present day section is from Cooper et al. (1995) and has not been modelled.
Fault dip angles close to 30°, as interpreted by Cooper et al. (1995) are reasonable for compression faults. Palaeogene shortening could be even less taking an average of dip values of 60 °f or former normal faults. An additional limitation of the Cooper et al. (1995) interpretation for the deep geometry of the faults is that the lower crustal detachment interpreted by these authors is deeper that the crustal thickness suggested by stretching models applied to the Mesozoic history of the area. In conclusion, it appears that all the limitations described for the model make it insufficiently accurate to predict details of the structure, palaeotopography or the stratigraphic relationships between Palaeogene stratigraphic units and related unconformities. However, an important result of the model is that the small amount of compressional shortening that occurred during Palaeogene was enough to generate the amount of subsidence/rock-uplift suggested by the Palaeogene sedimentary record and the inferred flexural-modelled palaeo topography.
12. DISCUSION 12.1. CONSTRAINING DIFFERENT TECTONIC SCENARIOS FOR THE PALAEOGENE To constrain the possible alternative tectonic scenarios on Palaeogene tectonic history, I used quantitative thermal subsidence and flexural models to test how plausible these different tectonic scenarios are. Although the observed subsidence could not be explained by any individual hypothesis, a
128
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
combination of all effects gave a satisfactory result. The observe d local flexural component of subsidence is interpreted as produced by uplift of local faulted block s (Fig. 3.1). Van der Hammen (1961) was the first geologist who recognized Palaeogene tectonic act ivity in the ar ea of the EC. This interpretation is supported by sedimentological (Anderson, 1970; 1972; Butler and Schammel, 1988; Hoorn, 1988; Laverde, 1989; Alfonso, 1989; Ecopetrol-ICP, 1996; Gómez et al., 1999), palynological (Van der Hammen, 1961; Sarmiento, 1992, 1993), petrographical (Porta, 1965; Anderson, 1970, 1972), mineralogical (Cheilletz et al., 1993, 1997; Branquet et al., 1996), structural (Morales et al., 1956; Julivert, 1970; Butler and Schammel, 1988; Guillande, 1988; Branquet et al., 1996, 1999b; Casero et al., 1995, 1997; Rathke and Coral, 1997; Corredor, 1997; Restrepo-Pace, 1999a,b; George et al., 1997; Branquet, 1999; Gómez et al., 1999) and fission track (Shagam et al., 1984; Toro, 1990; Van der Wiel, 1991; Van der Wiel and Andriessen, 1991; Hossack et al., 1999; Gómez et al., 1999) evidence for preAndean Palaeogene compression deformati on or slight local rock-uplift of the EC and MV. Incipient inversion of Mesozoic extensional basins (Fig. 3.1, e.g. Gomez et al., 1999) was probably the mechanism responsible for uplift of local faulted blocks.
12.2. PALAEOGENE TECTONIC HISTORY AND PLATE-TECTONICS Inverted extensional basins, upthrust basement blocks and whole lithospheric folds (Cloetingh et al., 1999) are common intraplate compressional/ transpressional structures that affect continental plate margins and/or plate interiors (Ziegler et al., 1998). Compressional features at plate margins are commonly associated with collision-related orogenic processes. Compressional features at plate interiors are usually not associated with orogeny. (Ziegler et al., 1995). In Colombia all plate-tectonic interpretations ( e.g. Pindell and Erikson, 1993; Pindell and Tabut, 1995) propose collision of the Caribbean with northwestern South America. Collision was oblique and diachronous, becoming younger northward (Pindell and Erikson, 1993; Pindell and Tabut, 1995). Ther efore plate-tectonic history suggests that Palaeogene basin inversion and upthrust of basement blo cks (Fig. 3.1) were collision-related, probably involved right-lateral transpressional deformation, and led to pre-An dean orogeny in the Central Cordillera during the Palaeogene. Transpressionally deformed grabens are also associated with zones of major wrench faulting (Ziegler et al., 1995). Probably some right-lateral strike-slip faults ( e.g.P alestina Fault, Irving, 1971) were active during the Palaeogene. Compressional/transpressional stresses related to collisional plate interaction are responsible for reactivation of pre-existing crustal discontinuities ( Fig. 3.1), upthrust of basement blocks and inversion of tensional hanging-wall basins (Ziegler et al ., 1995). According to Ziegler et al. (1998) these compressional structures can occur at distances up to 1600 km from the collision front, both in the forearc (foreland) and backarc (hinterland) positions with respect to the subduction system controlling the evolution of the corresponding orogen. Compression in the region behind a magmatic arc is associated with Andean-type orogens and occurs during periods of increased convergence rates between the subducting and overriding plates (Ziegler et al., 1998). Inversion of rift tensional hanging-wall basins located behind a magmatic arc is the result of acceleration of convergence rates between the colliding plates, their increased mechanical coupling and the transmission of compressional stresses into the backarc domain of the overriding plate (Uyeda and McCabe, 1983; Ziegler, 1993 in Ziegler et al., 1998). Rates of the Caribbean-South America plate convergence changed during the Cenozoic. Periods of development of compressional structures seem to correlate with times of high convergence rate, particularly during the Eocene (Daly, 1989) and during the Late Miocene-Pliocene, i.e. Andean orogeny (Cooper et al., 1995). Probably during late Oligocene-Early Miocene, development of compressional/transpressional structures was associated with the rupture of the Farallon Plate into the Cocos and Nazca plates about 25 Ma ago (Wortel and Cloetingh, 1981; Duncan and Hardgraves, 1984). Ziegler et al. (1995) cited several examples of intraplate compressional/transpressional structures in Europe developed during phases of plate boundary reorganisations that ultimately lead to the break-up of plate assembly. Build up of intraplate compressional/transpressional stresses in the NW margin of South America during latest Cretaceous-Palaeogene could have been favoured by subduction impediment
129
Chapter 3
caused by the arrival of more buoyant oceanic crust, such as an oceanic plateau. Nivia (1987), Kerr et al. (1996, 1997) and Sinton et al. (1998) based on geochemical evidence have proposed that the accreted oceanic terranes of western Colombia (Amáime Terrane in the werstern flank of the Central Cordillera, Calima Terrane in the Western Cordillera and Cuna Terrane in the Serranía de Baud ó, Fig. 1.1) are similar to the anomalously thick and buoyant Caribbean Plate. According to these auth ors the Carib bean Plate west of Colombia was young lithosphere and an oceanic plateau. Subduction processes can be impeded if buoyant material, such as a spreading ridge, oceanic plateau or micro-continent, collides with a mature arc-trench system (Muellert and Philips, 1991; Cloos, 1993, all in Ziegler et al., 1998). Large obstacles as oceanic plateaus can provide sufficient subduction resistance to deform the arc-trench, causing the build up of compressional stresses in the subducting plate and potentially its imbrication (Ziegler et al., 1998). This can explain intense deformation in the accreted oceanic plateau terranes of the Western Cordillera (Nivia, 1987). Build up of compressional/transpressional stress probably also favored deformation in the continental plate margin represented by the Central Cordillera. Ziegler et al. (1998) suggested that at any stage in a subduction zone the upper plate continental margin is weaker than the oceanic lithosphere plate margin. This suggests that the upper plate margin is the most likely candidate to be compressionally reactivated. If a magmatic arc was present in the Central Cordillera, as suggested by Palaeogene magmatic evidence, the Central Cordillera lithosphere was hot, weak and prone to deformation. According to Ziegler et al. (1998) for the build up of intraplate compressional stresses in forearc and foreland domains, the following collision-related scenarios are envisaged: (1) During the initiation of a subduction zone along a passive margin or within an oceanic basin. (2) During subduction impediment caused by the arrival of more buoyant crust, such as an oceanic plateau or a micro-continent at a subduction zone. This is applicable to Colombia during Late Cretaceous-Palaeogene time.
12.3. MECHANICAL ASPECTS OF BASIN INVERSION Mechanical aspects of basin inversion depend on the interplay of stresses and rheology of the lithosphere. 12.3.1. Str esses
We have discussed in terms of plate-tectonics the build up of collision-related intraplate compressional stresses at the NW South American plate margin. According to Ziegler et al.( 1998), this build up is indicative for mechanical coupling between an orogenic wedge and its fore- and/or hinterland (Ziegler et al., 1998). If this is applicable to the study area it would imply some mechanical coupling between the orogenic wedge, represented mainly the Central Cordillera and the regions east of it (MV, EC, and LLA) during Palaeogene time. However, the intensity of collisional coupling between an orogen and its fore- and hinterland is temporally and spatially variable. This can be a function of oblique collision (Ziegler et al., 1995, 1998). 12.3.2. Rheology of the l ith osphere
Localization of collision-related compressional intraplate deformation is controlled by spatial and temporal strength variations in the lithosphere, in which the thermal regime, the crustal thickness, the pattern of pre-existing crustal and mantle discontinuities, as well as sedimentary loads and their thermal blanketing effect play an important role (Ziegler et al., 1998). The strength of the continental lithosphere is controlled by its depth-dependent rheological structure in which the thickness and mineralogical composition of the crust, and the thickness of the mantle lithosphere. The latter depends largely on the potential temperature of the asthenosphere, as well as the presence of fluids and strain rates, play a dominant role (Stephenson and Cloetingh, 1991; Ranalli, 1995; Ziegler et al., 1995; Cloetingh and Burov, 1996).
130
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
Important inferences about lithosphere rheology in the EC during Palaeogene can be inferred from flexural modelling results. The parameter that characterizes the apparent strength of the lithosphere is the flexural rigidity, which is commonly expressed through the effective elastic thickness of the lithosphere. For an elastic plate the notion of the integrated strength is quite close to that of the flexural rigidity. Burov and Diament (1995) have shown that the effective elastic thickness of the continental lithosphere is dependent on: (1)
(2) (3)
(4)
The thermal state/age of the lithosphere (thermal age defined as a period of time required for the lithosphere to reach its present-day thermal state, assuming that the lithosphere was initially melted). The thermal age controls the depth to a specific geotherm obtained from a plate cooling model, assuming that the lithosphere did not undergo thermal re-setting during this time. The thermal age gives the age of the last large-scale thermal event (Burov and Diament, 1995). The thermal state of the lithosphere controls lithospheric strength, since temperature dependent creep controls the ductile strength of the lower crust and lower mantle lithosphere. Flexural models in the EC infer a weak lithosphere (EET 5 km in the area of the former Mesozoic extensional basin). In this area during Paleocene the thermal age of lithosphere was very young since the last stretching thermal event was Aptian (see Chapter 2) Weak lithosphere corresponds to thermally destabilized lithosphere. Young rifts and volcanic areas are characterized by low elastic thickness values (e.g. Ebinger et al., 1989, in Burov and Diament, 1995) due to thermal weakening and necking of the lithosphere. Stretched lithosphere is weak and low elastic thickness values in the EC may have inherited this weakness. This could explain the very low effective elastic thickness ( 5 km) in the area of the EC during Palaeogene time. However, the thermal anomaly associated with an active rift decays rapidly upon termination of crustal extension; after 60 Ma about 65 %, and after 180 Ma about 95 % of the anomaly had decayed (Ziegler et al., 1995). The coupling or decoupling state of the crust and mantle. The thickness and proportions of the mechanically competent crust and mantle. The thickness of the mechanically competent crust and the degree of coupling or decoupling are generally controlled by composition of the upper and lower crust, the total thickness of the crust, and by the crustal geotherm. If decoupling take place, as is of common occurrence in continental lithosphere, it permits as much as a 50 % decrease of elastic thickness, compared with elastic thicknesses implied for conventional thermal profiles (Burov and Diament, 1995). Low values of EET are usually associated with fault-controlled upper crustal flexure patterns (Van Wees and Cloetingh, 1994). In the EC area during Palaeogene the reduced EET values (<10 km) inferred from flexural models suggest that lithosphere strength was represented only by the upper crust strength (c.f. Cloetingh and Burov, 1996). This compressional intraplate deformation probably affected only the upper crust. Probably Palaeogene transpressional deformations in the area of the EC may have involved “simple shear” type detachment of the crust at the level of the rheologically weak lower crust from the mantle lithosphere, as suggested by Ziegler et al. (1995) in many other areas. The local curvature of the plate, which is directly related to the bending stress (Burov and Diament, 1995). The curvature of the plate depends also on the rheological structure and on the distribution of external loads applied to it ( e.g. topography, sediment fill and plate boundary forces). Bending stresses created by major mountain belts are large enough to cause inelastic deformation (brittle failure and ductile flow) in the underlying plate, which in turn, leads to a 30 to 80% decrease of elastic thickness beneath such belts (and less beneath the adjacent regions). The boundary forces and moments may lead to more localized but even stronger reductions in elastic thickness (Burov and Diament, 1995). Flexural models for the EC during Palaeogene infer small bending stress and small plate curvature. Thus this effect probably was not important.
Additionally the great thickness of sedimentary fill in the area of the EC, could have produced a blanketing thermal effect that inhibited lithosphere cooling, so preserving a relatively hot and weak lithosphere. Ziegler et al. (1998) have shown that the integrated strength of the lithosphere at a continental margin is markedly reduced if there is a thick syn- and post-rift sedimentary prism.
131
Chapter 3 12.3.3. Pre-exi stin g crustal disconti nu iti es
An additional important effect that controls lithosphere strength is the presence of pre-existing crustal discontinuities that weaken the lithosphere and play a crucial role in localising intra-plate compressional deformations. Standard rheological models that do not include pre-existing crustal discontinuities are unable to explain inversion of rift basins lacking a thick post-rift sedimentary prism (Van Wees, 1994; Ziegler et al., 1998). Deep reaching pre-existing crustal discontinuities, such as lithological inhomogeneities associated with ancient suture zones, as well as extensional, thrust and wrench faults, cause significant weakening of the upper crust. Such discontinuities are apparently characterized by a reduced frictional angle, particularly in the presence of fluids (Van Wees, 1994; Sibson, 1995; Ziegler et al., 1995). A major factor in the compressional reactivation of rifted basins is the failure of the upper crust along pre-existing faults causing stress concentration on the mechanically strong part of the upper mantle, resulting ultimately in its yielding. Reactivation of relatively steep dipping normal faults occurs when the angle between their strike and the compressional stress trajectory is smaller than 45 ° (Allemand and Brun, 1991; Ziegler et al., 1995). In the EC results of flexural models suggest that t he former Mesozoic extensional hanging-wall basins were slightly inverted during Palaeogene time (Fig. 3.1). This corroborates the importance of pre-existing crustal discontinuities to focus deformation. The reactivation of faults also suggests an oblique orientation of the compressional stress trajectories, which is in agreement with oblique plate convergence. In conclusion, incipient inversion of Mesozoic extensional basins and upthrusted basement blocks in the EC area was favored by collisional right-lateral transpresssional stresses that affected a weak lithosphere and reactivated previously existing discontinuities, i.e. the inversion of former Mesozoic extensional faults ( Fig. 3.1). The earlier deformation in the western MV (e.g. George et al., 1997; Restrepo-Pace et al., 1999a,b) compared to the EC or LLA foothills suggests that basin inversion “prograded” eastwards. This requires that basins located more proximal to the collision front were characterized by a mechanically-weaker pre-inversion lithosphere than more distal basins. It appears logical that the “weakest” basins are prone to early inversion. (Ziegler et al., 1995). The lack of important compressional intra-plate deformations in the LLA can be related either to the absence of major crustal discontinuities in the respective foreland, or to the lack of mechanical coupling between the LLA foreland and the orogen during the evolution of the latter (Ziegler et al., 1995). The Palaeogene stratigraphic record of the EC and neighbouring MV and LLA basins is useful to date collision-related intraplate compressional deformations affecting the northwestern margin margin of South America ( c.f. Ziegler et al., 1995).
12.4. LARGE-SCALE BASIN WIDE FLEXURE AND STRUCTURAL GEOLOGY For forward stratigraphic modelling of piggyback basins the importance of coupling large-scale basin wide flexure with thrusting on sub-basin scale and sedimentation has been demonstrated by Zoetemeijer et al. (1993). I have incorporated structural geology in basin modelling in order to bridge the basin-wide and sub-basin scales (Cloetingh et al ., 1994) applying the Thrustpack model. The results are consistent with the interpretation of incipient inversion of the hanging-walls of major Mesozoic normal faults during Palaeogene time ( Fig. 3.1). The results of the modelling approach as well as the increasing evidence indicates that an important co mpression/trans pression event took place related to inversion of the Mesozoic extensional basin in the area of the EC ( Fig. 3.1). Also the results of this study suggest that the Palaeogene deformation history of the E C is related to inversion processes mainly of the faults limiting the original Mesozoic extensional bas in ( Fig. 3.1). If this interpretation is true it implies a compressional stress-field oblique to the fault strike. This is consistent with oblique plate convergence with a right-lateral strike-slip component. The total amount of shortening was moderate (<10 km). Deformation started close to the Central Cordillera and prograded eastward.
132
Palaeogene Incipient Basin Inversion History of the Eastern Cordillera Colombian Andes
12.5. SUGGESTIONS FOR FUTURE STUDIES Although there is an increasing amount of evidence of earlier pre-Andean deformation in the EC, and there is an increasing interest in the study of the Tertiary sedimentary record of the EC, many uncertainties remain. Main uncertainties are related to the partial erosion of the sedimentary record and to age determinations and stratigraphic correlation. All these uncertainties make any attempt to use the available data difficult. In the case of this study I tried to compile the best possible database. Nevertheless some assumptions were necessarily applied to generate the palaeogeographic and thickness maps that constitute the conceptual model on which the modelling approach is based. An important constraint on the thermal, uplift and erosion history of any orogen is fission track data. However, the number of analyses available for the EC is very small and track lengths in most cases have not been neassured. Clearly they are inadequate. Understanding of the tectonic and sedimentary history of the EC needs more and better stratigraphic and fission track data. Palaeogene compressional deformation has been documented locally (Resterpo-Pace et al., 1998, 1999a,b) also for the emerald mineralizations of the EC (Branquet, 1999). However, clear and detailed identification of such Palaeogene deformational history and its structural style is difficult because it has been partially or totally masked by the later and stronger Andean deformation. More and better surface and subsurface data are also needed to understand Palaeogene deformation. Large volumes of Cretaceous organic-rich shales (Mayorga and Vargas, 1995; Mora et al., 1996, 1997; Mora 1997) were mature enough to generate and to expulse large volumes of hydrocarbons during Palaeogene time (Mora et al., 1996, 1997; Mora, 1997). The Palaeogene deformation history could have created structural traps for such hydrocarbon charge. Understanding of such Palaeogene deformation history, its structural styles and the preservation and modification of Palaeogene structures during Andean deformation could be a key element for further oil exploration.
13. CONCLUSIONS Flexural models and numerous evidence, such as the presence of Palaeogene unconformities, e.g. the regional Eocene unconformity (which locally truncates structures and other local unconformities), lateral changes of facies and thickness, local erosion indicated by detrital composition of sandstone and limited fission track data suggest that an incipient inversion of Mesozoic extensional basins occurred during Palaeogene time. Palaeogene basin inversion was related to the collision of oceanic plateau terranes with the northwestern margin of South America. Build up of intraplate compressional/transpressional stresses in the NW margin of South America during latest Cretaceous-Palaeogene could have been favoured by subduction impediment caused by the arrival of more buoyant oceanic crust, such as an oceanic plateau (Nivia, 1987; Kerr et al., 1996, 1997; Sinton et al., 1998). Right-lateral transpressional deformation likely lead to a pre-Andean orogeny in the Central Cordillera during the Palaeogene. Probably some right-lateral strike-slip faults ( e.g. Palestina Fault, Feininger, 1970; Irving, 1971) were active during Palaeogene. Periods of development of basin-inversion and compressional structures seem to correlate with times of high convergence rate, particularly during Eocene (Daly, 1989) and Late MiocenePliocene, i.e. during Andean orogeny (Cooper et al., 1995). During late Oligocene-Early Miocene, development of compressional/transpressional structures was associated with the rupture of the Farallon Plate into the Cocos and Nazca Plates about 25 Ma ago (Wortel and Cloetingh, 1981; Duncan and Hardgraves, 1984). Development of compressional/transpressional structures suggests some mechanical coupling between the orogenic wedge, represented mainly the Central Cordillera, and the regions east of it (MV, EC, and LLA) during Palaeogene time. However, the intensity of collisional coupling between the orogen and its fore- and hinterland is temporally and spatially variable, depending on the obliquity of the collision (Ziegler et al., 1995, 1998). Subsidence in the pre-Andean LLA, palaeo-EC and MV during Palaeogene time cannot be explained only by the topographic load of the palaeo-Central Cordillera as if the basin were a simple foreland basin (as interpreted in literature e.g. Cooper et al., 1995). Palaeogene subsidence only could be modelled assuming three different subsidence components: (1) Residual thermal subsidence after Mesozoic rifting; (2) Flexural subsidence of the lithosphere due to topographic load of the Central
133
Chapter 3
Cordillera; and (3) Flexural subsidence of the lithosphere produced by incipient topography generated during the Palaeogene. Flexural subsidence models assuming the lithosphere behaves as an elastic plate of laterally variable thickness, and including these three subsidence components along four regional cross-sections, suggests that during the Palaeogene local topography (up to 500 m) was developed close to the borders of the former Mesozoic extensional basin, probably by inversion along the border extensional faults. However, such topography was probably discontinuous and low. It did not significantly disturb the sedimentary and palaeocurrent pattern in the Palaeogene basin. Modelling results require very low values of effective elastic thickness in the area of the former extensional basin. The Mesozoic rifting events reduced significantly the strength of the lithosphere, making it very prone to Palaeogene deformation and to further Andean deformation. Low values of effective elastic thickness are usually associated with fault-controlled upper crustal flexure patterns (Van Wees and Cloetingh, 1994). In the EC area during Palaeogene the reduced EET values (<10 km) inferred from flexural models suggest that lithosphere strength was represented only by the upper crust strength ( c.f. Cloetingh and Burov, 1996). This compressional intraplate deformation was restricted to crustal levels involving “simple shear”-type detachment of the crust at the level of the rheologically weak lower crust from the mantle lithosphere. The great thickness of sedimentary fill in the area of the EC could have produced a blanketing thermal effect that inhibited lithosphere cooling, thus preserving a relative hot and weak lithosphere during the Palaeogene. Incipient inversion of Mesozoic extensional basins occurred by reactivation of previously existing discontinuities, such as Mesozoic extensional faults. The occurrence of earlier deformation in western the MV ( e.g. George et al., 1997; Restrepo-Pace et al., 1999a,b) as compared to the EC or LLA foothills suggests basin inversion “progradation”. This requires that basins located more proximal to the collision front were characterized by a mechanically weaker pre-inversion lithosphere than that of more distal basins. Structural kinematic modelling along a regional cross-section suggests that the amount of shortening during Palaeogene necessary to produce that Palaeogene topography is small and dependent of the dip angle of the Palaeogene contraction faults. Assuming a dip angle close to 30º and using the structural cross-section interpretation by Cooper et al. (1995) the modelled total amount of shortening at the end of Early Miocene appear to has been less than 10 km. A direct consequence of this interpretation is the possibility of generation of hydrocarbon traps during Palaeogene, a time when petroleum generation and migration occurred according to petroleum system modelling results published in the literature. Oil generation in Cretaceous rocks in portions of the EC happened a long time before formation of late Oligocene-Miocene anticlines, the oldest Tertiary traps preserved along the foothills (Gomez et al., 1999).
134
CHAPTER 4 NEOGENE BASIN INVERSION HISTORY OF THE EASTERN CORDILLERA, COLOMBIAN ANDES 1. INTRODUCTION Inverted extensional basins and upthrust basement blocks are common intraplate compressional/transpressional structures that affect continental plate margins (Ziegler et al., 1998). Compressional /transpressional stresses that are related to collisional plate interaction are responsible for inversion of tensional hanging-wall basins (Ziegler et al., 1995). Compression in the region behind a magmatic arc is associated with Andean-type orogens during periods of increasing convergence rates between the subducting and overriding plates (Ziegler et al., 1998). During basin inversion, basincontrolling faults reverse their movement due to compressional/transpressional stresses, and basins become positive features. The Neogene tectonic history of Colombia and particularly the history of the EC (Fig. 1.1) are an excellent example of complete inversion of a former extensional basin and formation o f a mountain range related to the Andean Orogeny. According to the current literature ( e.g.C ooper et al., 1995) the former Mesozoic extensional basin was inverted during Neogene time to form the EC, w hile the LLA and Middle MV became independent foreland basins related to flexural loading of the E C (Figs. 1.1 and 1.2). Up till now no flexural modelling has been carried out to test this hypothesis. A record of the tectonic rock-uplift and exhumation history of the EC has been preserved in the synorogenic sedimentary record of the neighbouring MV and LLA foreland basins. The study of the sedimentary fill of these basins gives important information about basin formation mechanisms ( e.g. Cloetingh et al., 1993) and about the tectonic rock-uplift of the EC during Neogene time. The aim of this Chapter is to contribute to understanding of the tectonic basin forming and inversion mechanisms during Neogene in terms of geodynamic processes that govern deformation of the lithosphere. This Chapter studies the tectonic subsidence of the MV and LLA Neogene basins. The hypothesis that tectonic subsidence of these basins was the result of lithosphere flexure produced by the increasing topographic load represented by the uplifting EC during Neogene time is successfully tested through modelling. The contribution of thermal subsidence after Mesozoic stretching is also tested for Ne ogene time. The surface-uplift evolution of the EC is delineated from flexural models (see location on Fig. 4.1) and its exhumation is evidenced by scarce fission track data from literature (Kohn et al., 1984; Shagam et al., 1984; Van der Wiel, 1991; Andriessen et al., 1993; Hossack et al., 1999). Finally the evolution of the lithosphere strength through Neogene time is delineated and discussed.
2. TECTONIC SETTING 2.1. PLATE TECTONIC INTERPRETATIONS 2.1.1. M iddl e M iocene
Collision and accretion of the Choco Block (Serrania de Baudo in Fig. 1.1, accreted oceanic Cuna Terrane according to terminology of Toussaint, 1995a,b) with the n orth-western margin of South America occurred during the Middle Miocene (Duque-Caro, 1990) and may have contributed to loading and have initiated deformation in the EC. An Andean foreland basin was developed since Middle Miocene in the area of the LLA (Cooper et al ., 1995). 2.1.2. L ate M iocene–Pliocene
The South American Plate rate convergence was fast, inducing deformation in the Colombian Andes (e.g. Cooper et al., 1995). Major deformation of the EC began at approxi mately 10.5 Ma (Cooper et al., 1995). During this deformation phase the EC was uplifted and erode d (Fig. 1.2). Pre-
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
1400
1300
Serrania de Perija
s a c u L n a S e d ia n 42 a + r r e 70 71 S +
y) r n d a u b o t e p la e r e 43 + ( F
72 +
Medellin
e s d A n a r i d e M
Cucuta
52 53 +
25 +
73 27 +
1200
Maracaibo Basin
26 -
23 -
0
Barinas Basin
100
200 Km
Bucaramanga 74 +
47 +
18 +
1
Arauca 44 +
r a i le rd o C
7 -
Llanos Orientales Basin (LLA) 2
1100
Tunja
Yopal
Manizales
1000
900
r e u t u S a l e r o m R
3 F i g .
Bogota Ibague
4 . 6
4
Villavicencio
Fig. 4.7
l n t ra e C
5 9 6
Neiva a e n e d r a a i c n a a M r r e a S L
800
800
900
1000
San Jose del Guaviare
1100
8 13 1200
10 1300
11 1400
12
7 1500
Figure 4.1: Location of stratigraphic columns and wells used (see Table 2.1) and location of lithospher e flexur e 2D model sections. Numbers along sections refer to labelling of stratigraphi c transver sal sections (Figs. 2.7, 2.8 a nd 2.9) and also to labelling of lithosphere flexure 2D models ( Figs. 4.10 to 4.12 and 4.14). The Romeral suture i s the westernmost boundary of continental crust. It was used a s a free plate boundar y f or the 2D flexural models.
existing extensional faults were inverted and new compressional structures developed. On the western flank of the EC and in the MV, middle Eocene folds were reactivated (Buttler and Shamel, 1988). Erosional deposits from the EC are preserved in the Guayabo FM in the LLA. Deformation and rock-uplift are still active, per iodically causing earthquakes. The Neogene sedimentary record of the Sabana de Bogotá ( Fig. 2.1) suggests that 1000 to 2000 m of surface-uplift occurred between 5 and 3 Ma (Van der Hammen et al., 1973; Van der Hammen and Hooghiemstra, 1997; Hooghiemstra, 1984, 1989; Helmens, 1988, 1990; Andriessesn et al., 1993). Present day plate velocity vectors measured from satellite data indicate that the Colombian Andean region (Andean Block) is moving northeastward with a right-lateral transpressional movement relative to the cratonic South American Plate (Freymueller et al ., 1993; see Chapter 5).
136
Chapter 4 MIDDLE MIOCENE
1400
PALEOGEOGRAPHY Positive relief Alluvial fan an d flu vial Coastal plain predominanatly sandstones Coastal plain predominanatly mudstones
1300
1200 u
1100
1000
B S u a c n a t a r a n m d e a r n M g a a s f a s if u lt
s a c u L n a S e d i a n r a r e S
Maracaibo Basin Cucuta
e s d A n i d a r M e 0
Barinas Basin
100
200 Km
y l e l a a V m l e n Arauca Littoral to inner shelf sandstones a e t d a n s e l Shallow marine inner shelf s y a g d a g M n carbonates t l a i Shallow marine inner shelf M a u a z o a s f o n e l i l mudstones and siltsones b b B o s a d s a n a i d m i l T u a B a C Outer shelf shales or carbonates M S t e d S ) o g i o n y S r g u Turbiditic sandstones C E L a n v e ( j a R e o c e m n I a s t u C y r T Tunja s a in a o v i l l e l t r a d s b m u n d I e r n ba a l m t e f o l o i a s C H b e o C y r d a u n Yopal l s a r o S t p a t m l t e C i u l Manizales s o u a u a t a f i a m B f E d e a o t in a r z a c n a I Llanos Orientales e r d a s b a g o t u a a a m G o ira b - b h S Bogota B C G u Bucaramanga
Ibague
900 l r a t n e C
800
800
S y e l a V n a l e t e m a a s e n l g d s y a a d M l t u a g f a M e r p t p u l U f a a i n v a a s m i r Neiva i e b N b - A l t a u S
900
Villavicencio
a e n e d r a i a n c a a r M r e a S L
1000
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 4.2: Middle Miocene palaeogeography without palinpastic restoration. (modified from Geotec, 1992 and Cooper et al., 1995).
3. STRATIGRAPHY In this section I summarize the Neogene stratigraph y (Figures 2.7 to 2.11). Because the deformation and rock-uplift history of the EC i s recor ded by the sedimentary record of the Sabana de Bogotá and the MV and LLA basins ( Figs. 4.2 t o 4.7), I also included additional geological evidence to constrain the Neogene tectonic history of the EC. The Middle Miocene to Present late Tertiary sedimentary record of the study area is present in the LLA and MV Neogene Andean foreland basins (Cooper et al., 1995,) and locally also in the Sabana de Bogotá (Fig. 1.2 a nd Figs. 4.2 to 4.7).
3.1. LLANOS ORIENTALES (LLA) According to Cooper et al. (1995), during Middle Miocene a global tectono-eustatic base-level rise coincided with the first significant deformation in the EC. Evidence for at least partial emergence
137
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
of the EC is the presence of more sand in the western foothills than in the east (Cooper et al., 1995). In the LLA foothills a moderate compressional tectonic phase occurred during Middle Miocene, ending
MIDDLE
s a c u L n a S e d i a n a r r e S
MIOCENE THICKNESS
(meters)
Medellin
B u Serrania c a r a de Perija m a n g a
Bucaramanga
Tunja
o b a m m e a s t C y a t s l m u i u a t f i B Bogota
n a l e t e m a d y s a g s M l t u f a
l n t ra e C Neiva
800
800
t u l f a i r a m t a A l
900
e a n d e r a c a i n a a r M r e a S L
1000
ra a m m á te T y s a s n ia lt s u u a C f
o a e l p a m o s t e a m s y r t a l i c a u a f u G
t u l f a a e z m h a C
Barinas Basin
Arauca
a m i n t e l s a S s y a L l t u a f
Ibague
900
e s d A n a r i d e M
Cucuta
f a u lt
a l e r i o rd C
Manizales
Maracaibo Basin
Llanos Orientales Basin (LLA)
Yopal
Villavicencio
0
100
-
200 Km
200
San Jose del Guaviare
1100
100 0
1200
1300
1400
1500
Figure 4.3: Middle Miocene restored thickness (meters) without palinpastic restoration. Thick lines represent palaeo-faults believed to be active during Middle Miocene time.
the previous monotonous sedimentation (Casero et al., 1995, 1997). The resultant loading tecto nically enhanced the h ighstand system tract and resulted in deposition of the green to dark grey Leon Fm ( Figs. 2.7 to 2.11 a nd Figs. 4.2 an d 4.3, Cooper et al., 1995) shales with iron nodules and locally organic-rich shales, deposited in anoxic conditions (Moreno and Velázquez, 1993). In the upper part an upward colour change from grey to red has been interpreted as a change from marine to continental deposition (Cooper et al., 1995). Later (since Late Miocene?) the coarse synorogenic continental clastic de posits of the Guayabo Fm were deposited in the LLA basin ( Figs. 2.7 t o 2.11 a nd Figs. 4.4 an d 4.5 t o 4.7). These sediments recorded the exhumation of the EC as evidenced by clasts of Cretaceou s rocks (clast s of mudstones, cherts) derived from erosion of the EC (Moreno and Velazquez, 1993). These authors recognized an erosional unconformity between the lower and middle portions of the Guayabo Fm. in the LLA foothills (near Nunchia).
138
Chapter 4
Seismic lines across the LLA foothills ( Figs. 2.1 and 4.7) show indications of severe thrusting of post-Leon age that seem to be overlain by Guayabo beds (Casero et al., 1995, 1997). Seismic (Figs. 4.6 a nd 4.7) and well data indicates a gradual westward increase of thickness to a maximu m in the thrust f aults defining the eastern boundary of the EC, and a virtually undeformed Neogene section on most of the LLA area ( Figs. 4.3, 4.5 a nd 4.6). However, in the south-western LLA, NE of the Serranía de La Macarena, seismic lines (e.g. Fig . 4.7) show that the upper part of the Neogene section (Guayabo Fm) has been partially eroded. This e xplains the decrease in preserved thickness in this part of the basin. Ecopetrol and Beicip (1995) have interpreted in this area some flower structures probably related to the Andean deformation. Wrench faulting in the LLA also has been proposed by Cediel (1982). Vásquez (1988) has interpreted a strike-slip component in several faults in the LLA Basin. Development of local rock-uplift associated with development of positive flower structures may explain the erosion of the Guayabo Fm in this area. It is interesting to note that southward in the Putumayo Basin (Fig. 1.1) the thinner Ospina Fm is considered time equivalent to the Guayabo Fm (but palaeontological arguments are weak) while the eastern flank of the Cordillera in that area is characterized by dextral strike-slip Altamira/Algeciras faults (Casero et al., 1995, 1997). LATE MIOCENE 1400
PALEOGEOGRAPHY Positive relief Alluvial f an and f luvial
1300
Coastal plain predominanatly sandstones Coastal plain predominanatly mudstones Littoral to inner shelf sandstones
1200
Shallow marine inner shelf carbonates Shallow marine inner shelf mudstones and siltsones Outer shelf shales or carbonates Turbiditic sandstones
1100 Manizales
1000
a in d s n ba o H b u S
o t in r a rd a s e l l b i ira b r d G u o Ibague S C y l l e r a t a n V e a C n e
a o b m m t e a s C y a s t m l i u t f a u i B
Bogota
n a l e t e m a d y s l a a g s d M l t a g u M f a r p e l t u U p f a i r a m l t a Neiva n A i a i v a s e b N b u S
900
800
800
Serrania Maracaibo de Perija Basin S B u a n c t a Cucuta a n r a d m e r a n g M a a s s if f a u lt
s a c u L n a S e d i a n a r r e S
900
e s d A n a r i d e M
Bucaramanga
Arauca
n a m e l ) t e C d a s E g y ( a s M n a t n l r i i u e o s s l z a i a f a l a B d B b a o r a b b n i C T u u l S a S r n S t e d r y t e a e s u v L a c n I o E
t u l f a a d e e z a a n m t b a o g o C h a a S B
1000
Llanos Orientales Basin (LLA)
e m s t y Yopal s l t u f a o a e l a p o m a r c á i a u G
Villavicencio
a e n d e r a c a i n a a r r M e a S L
r a a m e m a s t T y a s n i a lt s u u f a C
C v n I
Tunja
Barinas Basin
0
100
-
200 Km
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 4.4: Late Miocene palaeogeography without palinpastic restoration. (modified from Geotec, 1992 and Cooper et al., 1995). Detrital source areas. Clast
composition of the Guayabo Fm conglomerates changes upward. The lower part contains clasts of mudstone (one Oligocene clast as dated by palynology) and muddy
139
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
sandstones with coal fragments derived from the Carbonera Fm. The upper part contains clasts of chert or clasts with glauconite derived from Cretaceous rocks of the EC (Moreno and Velazquez, 1993). The coarsening upward trend reported for the Guayabo Fm (Ecopetrol and Beicip, 1995) might represent an increase in grain size of clasts derived from a growing topography of the EC.
3.2. MAGDALENA VALLEY (MV) Most of the Neogene record is continental ( Figs. 2.7 to 4.2 a nd 4.4). Thicknesses of Neogene strata are highly variable (Fig. 4.5). In the Middle MV the major units ar e the R eal Gp and the Mesa Gp In the Upper MV and the Honda sub-basin the major units are the Honda Gp and the Mesa G p ( Figs. 2.7 to 2.11). Although the general lithology and relative stratigraphic position of units with th e same name in different areas of the MV is similar, there is no evidence that these units are time equivalents or that they represent the same sedimentary events. Miocene deposits unconformably truncate early structures, evidencing deformation and erosion (Morales et al., 1956; Guillande, 1988). According to Jordan and Gomez (1996) the bases of the Honda and Mesa Gps vary spatially from conformable to angularly unconformable. This geometric variability, along with strong thickness and facies variations, suggests both palaeogeographic and tectonic controls on their accumulation. LATE
1400
B u Serrania c a de Perija ra 0 0 m 0 5 0 0 a n 0 g 1 0 a 0 5 f a 1 u l t 0 0 0 2
s a c u L n a S e d i a n a r r e S
MIOCENE QUATENARY THICKNESS (meters)
1300
0 5 0 0 0 0 0 1 5 1
u
0 0 1 0 0 0 0 0 5 0
1100 Manizales
5 1
4000
a o b m m t e a s C y a s t m l i u t u f a i B
72 0
Ibague
1000
Bogota
n a l e t e m a d y s a g s M l t u f a
900
Neiva
800
800
l t u f a i r a m l t a A
900
Barinas Basin
10 0
200 Km
Bucaramanga
0 0
1200
e s d A n i d a r M e
Cucuta
0
0
Medellin
Maracaibo Basin
a m i n 0 0 t e l s a S s y L a l t u f a
Tunja
t u l f a a e z m h a C
Arauca 4
0 0 5 3
o a m e m r a t i c y s u a l t s G a u f
C
0 0 0 3 0 0 0 0 2 5 2 0
0 0 5 1
Yopal
Llanos Orientales Basin (LLA) 0 0 0 1
0 5 0
Villavicencio
a e n d e r a i a n c a a r M r e a S L
1000
San Jose del Guaviare
1100
1200
0
1300
1400
Figure 4.5: Late Miocene restored thickness (meters) without palinpastic restoration. Thick lines represent palaeofaults believed to be active during Late Miocene time.
140
Chapter 4
0
1
2
3
E S
1 a i l u r a B o n a 0 C
1 5 8 C
m k 5
1 o t i h c i r o M 1 a n e m i r a u G
5 8 2 1 0 8 C
3 5 5 1 0 9 E R
a 2 4 4 8 N A S
1 a t n u P a L
m k 5
1 o r u a t n e C 1 a i r o l G a L 1 a n a b a C a L 1 a i r a M a L
m k 5
m F a r m e F n o b 5 4 3 2 1 n r a o o o o o o e b b b b b L C a a a a a p p y y y y y a a a a a o o u u u u u T T G G G G G
m F r o d a r i M p o T
c i o z e a l a P p o T
t n e m e s a B p o T
m k 5
3 1 9 R A
W N
m k 5
0
1
2
3
4
141
5 c e s
s r o t c e l f e r s r e k r a m c i h p a r g i t a r t s e h t g n i w o h s n i s a B s e l a t n e i r . O 1 . s 4 o n e a r l u L i g e F h t n i h g n u w o o r h t s e n n i l i o c t i c e m s s i e e s h t l f a o n n o i o i g t e r a a c o f o L . g n n o i i l t a l e t d e r o p r m e l t n a r I u : x e 6 . l f 4 n e i r u d g e i F s u
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes 3.2.1. Mi ddle M agdalena Val ley
Real Gp.A ccordi ng
to Cooper et al. (1995) the Real Gp lies unconformably over the Colorado Fm ( Figs. 2.7 to 2.11). They assume a Middle Miocene hiatus and stated that such an unconformity is evidence for deformation, exhumation and erosion of the EC. However, the age of the unit has not been confidently determined. Morales et al. (1956) reported a Miocene age based on fos sil leaves and a horizon of small gastropods found in the Velasquez field. The Real Gp consists o f conglomerate, sandstone and minor mudstone, representing alluvial fan and braided fluvial depositio n ( Figs. 4.2 an d 4.4, G eotec, 1994). Its maximum thickness is 3600 m near the EC but it thins westwar d (Fig. 4.5). Lateral thickness variations in the Middle MV, particularly a minimum of thickness in the Cachira palaeo-high (Fig. 4.5) suggest less subsidence in this area that was deformed simultane ously during deposition. According to Geotec (1994), piggyback deposits are common. Some faults affecting the Neogene sedimentary record confirm Neogene deformation. (Porta 1965, 1966; Servicio Geologico Nacional, 1967a,b; Ward et al., 1973; Restrepo-Pace et al., 1999a,b). In the Honda sub-basin the Honda Gp. consists of greenish gay sandstones (Wellman, 1970). From this Gp Gómez (1999) reported radiometric and volcanic fission track ages between 11.92 ± 0.11 Ma and 7.5 ± 0.6 Ma. Detrital source areas. Sandstones are characterized by a predominance of plagioclase and volcanic rock fragments with more plagioclase than K-feldspar and non undulatory quartz, pyroxene and other mafics as secondary components (Hathon and Espejo, 1997). Morales et al. (1956) reported hornblende, augite and a large amount of igneous and volcaniclastic material. These sandstones fall into the magmatic arc provenance field in a quartz, feldspar, lithic plot (Dickinson, 1985). Rare plutonic grains suggest a dissected arc provenance. The fresh volcanic fragments indicate a near detrital source infer red to be the Miocene age volcanics that blanket partially the Central Cordille ra ( Figs. 1.1 and 4.2; Hathon and Espejo, 1997). Palaeocurrent data also suggest dominant eastw ard flow of sediments (Rubiano, 1998). There is also evidence of detrital source areas located in the EC. Conglomerates contain black chert, mudstones and coal fragments (Morales et al., 1956) possibly derived from the EC. According to Ecopetrol et al. (1994) palynological evidence suggests these sediments were sourced predominantly from the EC. Mesa Gp. This unit lies unconformably over Real Gp ( Figs. 2.7 t o 2.11; Cooper et al., 1995). In the Honda sub-basin the Mesa Gp rests with angular unc onformity. Its age is Early Pliocene in the Honda sub-basin as determined radiometrically by Thouret (1988). In the Honda sub-basin its age is Early Pliocene (radiometric age 3.5 to 4.3 Ma: Thouret,1988; palynology: Dueñas and Castro, 1981). It consists of lithic conglomerates, sandstones and andesitic tu ffs containing pyrocl astic fragments. These sediments were derived from the Central Co rdillera and EC ( Figs. 1.1 a nd 4.5) and were deposited and reworked in alluvial fans and fluvial systems ( Fig. 4.5). Thickness reaches up to 575 m. (Morales et al., 1956). Porta (1965, 1966) has reported occ urrence of the Pliocene Mesa Fm, limited towards the east by the Honda Fault, indicating fault-controlled sedimentation. 3.2.2. Upper M agdalenaVal ley
According to Geotec (1994) sedimentation of the Middle to Upper Miocene Honda Fm shows a piggyback basin fill relationship with the west-verging thrusts that delimited the western flank of the EC and the Upper MV (Prado, Agua de Dios, Chaparral, Pitalito, Garzon faults). In the northern Girardot sub-basin Amezquita and Montes (1994) and Amaya and Santamaria (1994) demonstrated the activity of both an east- and west-verging fold-and-thrust systems. Both systems were active before and after sedimentation of the Middle to Upper Miocene Honda Fm. Lateral changes of thickness of the Middle to Upper Miocene Honda Gr also s uggest piggyback sedimentation. In the southern Neiva sub-basin ( Figs. 2.1, 4.2 a nd 4.4) sedimentation occurred on distal volcanic apr ons derived from the Central Cordillera volcanic arc, an d braided river systems flowing toward the east (Figs. 4.2 a nd 4.4, Van der Wiel, 1991). In the Neiva sub-basin the age of the Honda Gp is latest early Miocene to late-Middle Miocene (16.1 to 11.5 Ma), based on palaeontologic (Villarroel and Guerrero, 1984; Busbey, 1986; Czaplewski, 1989; Hirsfeld and Marshall, 1976;
142
Chapter 4
0
1
2
3
4
5
E m k 0 2
s o t i r g e N
0 1
0
1 - y a i p A 1 - a i r u S
d a t r e b i L a L
?
) . s m F ) . a r s e t m n n o F b e r o a m b e C a s y d a a n b u a G r e n i n a l d o i a n d i c t a a s r v i i r n M o C d o ( r e n L O s ( e a i n u n r o e e g e i a b n o c r m e a b a g e t a m c o l e a e r a e r N P C C P
1 - e n e m i h c i h C 1 - a e d a m u H
t l u a F a i l a s e T t l u a F a i r a M a t n a S
W
0
1
2
3
4
5
) C E S ( T W T
143
a l o r e f d ( a 1 i . n 4 a r e r e r u S i g e F h t n i f o n h w t r o o h n s n n i o s t a i B c e s s e o h n t a l f L o e n h o t i t f o a c o t r L a . p p u n r e o r h t G u o o s b e a y h t a u h G g e u o e r n t g o n o e i N t c e e s h t c f i o m s n i o e s i s o l r a e n l o i a i g t e r r a a p d . f o n a ) 5 n l 9 o a 9 i c t 1 a o , t l p e i r e c p h t i r e e t e B t n o d I n : N a . l 7 a . o 4 n r e t r e e r a p u c g a o c i F M E
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
Setoguchi and Rosenberg, 1985, 1987; Kay et al., 1989), radiometric and magnetoestratigraphic data (Hayashida, 1984; Guerrero, 1993). It consists of grey to tan feldspatic sandstone and drab to red brown mudstone and conglomerate, locally more than 1600 m thick. Conglomerates and sandstones contain abundant volcanic fragments, feldspar, quartz, chert, metamorphic and plutonic rocks. During deposition dacitic to andesitic volcanic activity increased in the north part of the Neiva sub-basin (Wellman, 1970). The Huila Gp (former Mesa Gp proposed by Howe, 1974 in the Upper MV; Guerrero, 1993) in the Neiva sub-basin it is of Late Miocene age, based on absolute ages (Takemura and Danhara, 1983; Setoguchi and Rosenberg, 1985; Van der Wiel and Van der Bergh, 1992a,b; Guerrrero, 1993). The lower part of the Huila Gp in the Neiva sub-basin (Neiva Fm; Guerrero, 1993)) consists of conglomerates with clasts of intermediate volcanic rocks, locally with abundant granitic pebbles. Volcanic arenite is the dominant sandstone with plagioclase. There are tan coloured mudstones. Detrital composition and palaeocurent data suggest these volcaniclastic sediments were mainly derived from the Central Cordillera, with some local exhumation of the Garzon Massif (Howe, 1974). The Huila Gp rests on angular unconformity on the Honda Gp (Guerrero, 1993) The Honda and Huila Gps were deposited in a piedmon t plain by coalescing alluvial fans and braided to meandering fluvial systems (Figs. 4.2 and 4.4), r elated to pulses of exhumation of the Central Cordillera. These units also re cord andesitic to dacitic volcanism of the Central Cordillera. The record of volcanism is best preserved in the Huila Gp (Van Houten and Travis, 1968; Howe, 1969; Van der Wiel and Van der Bergh, 1992a,b; Guerrero, 1993). Very coarse Pliocene to Recent clastic deposits are found along the western and eastern margins of the basin. The Upper Cenozoic record (Ibague Fan, Chaparral Fan, Ceibas Formation and Gigante Fm.) contains abundant volcaniclastic deposits derived from stratovolcanoes on the Central Cordillera. About 7 to 5 Ma ago several polymictic debris-flows and torrential sediments formed fans that spread across the Upper Magdalena Valley (Van Houten, 1976). In the southern part of the Upper Magdalena Valley changes in palaeocurrents and facies show that the exhumation of the Garzon Massi f (Fig. 2.1) began during accumulation of the upper part of the Honda Gp (late Middle Miocene) and continued during deposition of the Mesa Gp (Van Houten and Travis, 1968; Van der Wiel and Van der Bergh, 1992a,b; Guerrero, 1993). As mentioned in Chapter 3 w estward palaeocurrent directions in the Villavieja Fm (upper part of Honda Gp of the Upper MV) indicate that the southern EC was already a mountain range that completely closed the basin in its eastern side at about 11.8 Ma (ages from 40Ar/39Ar dates and magnetostratigraphy, Guerrero, 1993). Changes of these palaeocurrents indicate that exhumation of the southern EC began at 12.9 Ma (Guerrer o, 1993). Fission track data indicate that around 12 Ma. the Garzon Massif and the Central Cordiller a ( Fig. 2.1) wer e simultaneously exhumated. Such rock-uplift and exhumation caused rapid relative sub sidence of the intermontane Neiva sub-basin and deposition of increasingly coarse detritus from the wes t (Van der Wiel, 1991). After 12 Ma tectonic activity of both the Central Cordillera and the Garzon Ma ssif (Fig. 2.1) waned and sediments began to exhibit eastward and northward palaeocurent directions (Van der Wiel, 1991). Volcanic activity and exhumation of the Central Cordillera was resumed 10-9 Ma at the start of deposition of the Gigante Fm (Neiva member). Exhumation of the Central Cordillera occurred in its southern part where it merges with the Garzon Massif (Figs. 1.1 an d 2.1; Van der Wiel, 1991). Between 8 and 6.4 Ma activity of the Central Cordillera volcanic arc reached a maxim um and because of the high input of volcanic and volcaniclastic material into the basin, the north-flowing palaeo-Magdalena River was forced by volcanic aprons (Los Altares member of Gigante Fm; Van der Wiel, 1991) to flow along the easter n border of the basin. Around 6.4 Ma ago a new pulse of tectonic exhumation affected the Garzon Mas sif (Fig. 2.1), which supplied an increasing amount of metamorphic pebbles and boulders into the basin (Garzon Mem ber of the Gigante Fm) although sediment also came from the Central Cordillera. Rock-uplift of the massif culminated 6 Ma ago (Van der Wiel, 1991). During rock-uplift strong SE-NW compression produced folding and reverse faulting in the basinal sediments. Later sediments (lower unit of the Las Vueltas Fm) were immediately deformed after deposition, fo rming a small angular unconformity with older deposits. Rock-uplift of the Garzon Massif (Fig. 2.1) probably occurred along several faults of the Garzon-Suaza fault system. The latter had intermittent pulses of strike-slip movement, leading to the formation of several generations of
144
Chapter 4
alluvial fans coming from the east. Strike-slip activity of t he Garzón-Suaza fault system affected Quaternary sediments in local basins on the Garzón Massi f (Fig. 2.1). In the basin Quaternary sediments have been affected by reverse faulting, indicating NW-SE com pression (Van der Wiel, 1991). 3.3. SABANA DE BOGOTÁ The Sabana de Bogotá ( Figs. 2.1, 4.4 a nd 4.5) is a high plain in the axial part of the EC at an altitude of 2600 m. It is an intermontane basin c onsistent in a broad NNE-SSW-striking synclinorium probably formed during Oligocene to Late Miocene time. During late Pliocene and Pleistocene approximately 600 m of mainly lacustrine sediments accumulated (Helmens, 1988, 1990; Andriessen et al., 1993). The Neogene and Quaternary fluvial to lacustrine sedimentary section of the Sabana de Bogotá registers major tectonic surface-uplift for the period between 5 and 3 Ma (Helmens, 1988, 1990; Andriessen et al., 1993). A summary of this stratigraphy is the following: The Neogene-Quaternary stratigraphic record of the Sabana de Bogotá area starts with the Late Miocene Marichuela Fm, which represents very large synorogenic debris flows and gravity flows that aggraded on broad alluvial plains and lakes deposited during a period of increased regional tectonic activity, with locally strong deformation of strata (Helmens, 1990; Andriessen et al., 1993). The earliest Pliocene Tequendama Member of the Lower Tilata Formation represents a period of relatively quiet fluvial sedimentation. Palynological data indicate forest vegetation of tropical lowland type with an altitude of deposition that did not exceed 500 m (Helmens, 1990; Andriessen et al., 1993). Tilata Fm rests with a pronounced angular unconformity over a variety of older strata, confirming that some deformation preceded its deposition (Cooper et al ., 1995). The fluvial and alluvial fan sediments of the early Pliocene Tibagota member of the Lower Tilatá Fm were deposited at the beginning of the final major upheaval of the EC. Palynological and macrobotanical data indicate a depositional environment in the lower tropical to lower sub-Andean forest belt, between 1000 m and 1500 m (Helmens, 1990). The Late Pliocene Guasca Member of the Upper Tilatá Fm recorded fluvial to lacustrine sedimentation in the outer valleys of the present high plain area. Palynological data from that unit indicates deposition in the upper sub-Andean forest belt, at an elevation of about 2200 m (Helmens, 1990; Andriessen et al., 1993). Magnetostratigraphic data indicates that the Guasca Member was deposited between 3.2 Ma and 2.58 Ma (Helmens et al., 1997). Later the Late Pliocene unnamed Upper Member of the Upper Tilata Fm recorded sedimentation in the central part of the present high plain of Bogotá, when the main surface-uplift of the Sabana de Bogotá area had ceased. The Tilata Fm and correlative sediments are found on the slopes surrounding the Sabana de Bogotá, and their beds are often tectonically disturbed (Helmens, 1990). During the Early Pleistocene, lacustrine fluvial deposition of the Subachoque Fm occurred in a large area under alternately ‘glacial’ and 'interglacial’ conditions at an altitude similar to the present day elevation of the Sabana de Bogotá plain (which is 2600 m). The Subachoque Fm and younger stratigraphic units are tectonically undisturbed (Helmens, 1990; Andriessen, 1993). Later deposition of Pleistocene Sabana Fm in the central part of the present high plain of Bogotá occurred in a lake environment (Helmens, 1990; Andriessen et al., 1993). Palynological evidence indicates the replacement of a tropical lowland flora by a high mountain flora occurred during the Pliocene. This trend in vegetation change has been interpreted as being primarily the result of tectonic surface-uplift (Van der Hammen et al., 1973). Wijninga and Kuhry (1993), however, note than warmer climate in the Pliocene should have placed the altitudinal vegetation belts several hundred metres higher. Probably surface-uplift took place in various phases. Phases of high tectonic activity, with coarse gravity flow deposition alternated with phases of fluvial lacustrine sedimentation and soil formation (Helmens, 1990).
4.
TECTONIC SUBSIDENCE DURING THE NEOGENE
The study of the sedimentary fill of a basin and its associated tectonic subsidence signal give important information about basin formation mechanisms (Cloetingh et al ., 1993). In order to quantify the tectonic component of subsidence of the Neogene basins of the study area, a 1D-backstripping
145
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
technique was used (Steckler and Watts, 1978; Bond and Kominz, 1984). Details of the backstripping subsidence analysis, the procedure and data are described i n Chapter 2. The late Tertiary sedimentary record of the study a rea is present in the LLA and MV basins and locally in the Sabana de Bogotá. Because the Neogene sedimentary record of the LLA and Middle MV has not been studied, there are no reliable age data and I only used a limited number of sections.
4.1. RESULTS 4.1.1.
L lan os Or ientales (L L A)
The convex upward shape of tectonic subsidence cur ves (Fig. 3.16) clearly shows the acceleration of tectonic subsidence in a typical subsidence pattern, as observed in foreland basins (Allen and Allen, 1990). Maximum increase in tectonic subsidence occurred during the Neogene (last 10 Ma, see Fig. 3.16) during the Andean orogeny. The amount of tectonic subsidence is maximum (up to 1500 m, in well Arauca-1) near the thrust front of the Cordillera and gradually decreases eastward (e.g. well L a Tortuga-1 450 m, Fig. 3.16). 4.1.2.
M agdalena Val ley (M V)
The thickness distribution ma p (Fig. 4.5) indicates a general increase in thickness toward the EC in a similar pattern to forelan d basins. However, some lateral thickness variations in the Middle MV, particularly a minimum of thickness in the Cachira palaeo-high (Fig. 4.5), suggest that less subsidence took place in this area that, which was deformed simultaneously during deposition. Curves of tectonic subsidence ( Fig. 3.16) show large variability. Some curves show a convex upward shape with acceleration of tectonic subsidence especially during Neogene time, as observed in foreland basins (Allen and Allen, 1990). At other localities tectonic subsidence and sedimentation were compensated by local erosion resulting in a horizontal segments in the tectonic subsidence curves. Upper Magdalena Valley. Lateral changes of thickness of the Middle to Upper Miocene Honda Gr Also confirm piggyback sedimentation. In the Neiva sub-basin ( Fig. 2.1) subsidence in the area east of the Chusma thrust system started during the Early Miocene, probably more than 16 Ma with deposition of the Honda Fm just after the last phase of activity of the Chusma thrust system. The western limit of the subsiding area was the easternmost Dina thrust. Further to the east, the subsiding area was p robably bordered by very low isolated hills represented by erosional remnants of the Garzon Massif ( Fig. 2.1), which had been exhumated about 100 Ma ago. Between <16 Ma and approximately 12 M a subsidence was not yet very large (Van der Wiel, 1991). Tectonic subsidence curves of the Upper MV also show variability suggesting tectonic deformation and emergence of local blocks due to thrusting, with local erosion and simultaneous sedimentation in other parts of the area. Results of thermal modelling of vitrinite reflectance data (Buitrago, 1994) suggest that instead of flat subsidence curves, sedimentation and later erosion affected some local thrust faulted blocks. This latter case of subsidence and sedimentation was compensated by local erosion and indicates local exhumation and erosion of blocks, this suggests that tectonic defor mation affected some blocks while other areas were subjected to sedimentation. Figure 4.8 sh ows a map of the observed tectonic subsidence during Neogene time. This figure and the thickness ma ps ( Figs. 4.3 a nd 4.5) indicate subsidence in the Neogene LLA and MV basins and absence of subsidence in the area of the EC. Middle Magdalena Valley.
5.
NEOGENE REMAINING THERMAL SUBSIDENCE AFTER MESOZOIC RIFTING
In Chapter 3 I concluded that during Palaeogene time there was a partial component of subsidence produced by thermal equilibration of the lithosphere after the Mesozoic extensional events. I also tested through modelling the possibility of some Neogene remainin g thermal subsidence after Mesozoic lithosphere extension. Here I used the same methodology applied to Chapters 2 and 3: I
146
Chapter 4
calculated the remaining thermal subsidence due to thermal re-equilibration of the lithosphere during Neogene. A descr iption of the forward modelling technique and procedures developed by Van Wees et al. (1998) is given in Chapter 2. I use the lithosphere stretching factors calculated for Mesozoic rifting episodes (see tables Chapter 2).
5.1. RESULTS Results of these calculations appear in Figure 4.9. Comparison of the calculated thermal subsidence during Neogene time ( Fig. 4.9) with the observed tectonic subsidence ( Fig. 4.8) shows that the latter cannot be explained b y thermal re-equilibration of the lithosphere. Subsidence maps show that the remaining thermal subsidence assuming tectonic quiescence would have affected the area of the EC, where the Mesozoic extensional basin was located. The observed tectonic subsidence is just the opposite with zero values in the area of the EC, and maximum values in the Neogene LLA and MV basins. This opposit e trend is also clearly seen comparing the thick ness maps of the Lower Cretaceous (Figs. 2.13, 2.15 an d 2.17) and the Neogene (Figs. 4.3 a nd 4.5). These maps illustrate that the Mesozoic extensional basin located mainly in the area of the EC was completely inverted during Neogene t ime to form the mountain range. The Neogene sedimentation of the Saban a de Bogotá area ( Figs. 4.4 a nd 4.5), which apparently is an exception, just recorded the surface-uplift history of the Cord illera. s a c u L n a S e d i a n rr a e S 88.00
BACKSTRIPPED MIDDLE MIOCENE QUATERNARY SUBSIDENCE (m)
B u Serrania Maracaibo c a r a de Perija Basin m Cucuta a n g a f a u lt
e s d A n a r i d e M
0.00
0.00 440.00
Bucaramanga
160.00
ra a m m 520.00 á t e T y s s na ia lt 509.00 u us a C f 457.00
900.00
67.00 0.00
0.00
0.00
0.00 0.00
Medellin
0.00
r a i le d o r C
0.00
0.00
o 0.00 a b m m t e a s C y a s t m l i u u t f a i B
eo a la p m 0.00 o t e am s y s á r lt ai c a u u f G
0.00 1.00
0.00
Tunja
0.00
485.00
Manizales
0.00
0.00
0.00 -1669.00
0.00
0.00
-1505.00
Ibague
1.00
1.00
167.00 153.00
0.00
900
l n t ra e C
495.00
489.00
800
489.00
900
646.00
318.00
Yopal 497.00 l t u 327.00 726.00 516.00 f a 332.00 658.00 0.00 349.00 105.0 528.00 369.00 343.00 a 424.00 360.00 0 373.00 538.00 e z 641.00 399.00 270.00 m 467.00 354.00 337.00 a 370.00 478.00 472.00 325.00 C h 283.00 501.00 492.00
0.00
Villavicencio
0.00
294.00
288.00
414.00
306.00
275.00
216.00
323.00 300.00
0.00
167.00
208.00
141.00
254.00 196.00
240.00 548.00
0.00
Llanos Orientales Basin (LLA)
366.00
0.00
0.00
183.00 248.00 53.00 62.00 0.00
l t u a f 178.00 215.00 233.00 216.00 i r a 216.00 0 m Neiva l t a A 356.00
800
Bogota
0.00
169.00
638.00
0.00
201.00
485.00
0.00
0.00
Arauca
472.00
134.00
0.00
Barinas Basin
169.00
219.00
295.00
178.00 177.0 0 170.00
224.00
130.00
a e n d e r a a c i n a a M r r e a S L
1000
107.00
San Jose del Guaviare
1100
1200
-
0
100
1300
200 Km
1400
1500
Figure 4.8: Observed tectonic subsidence, in meters, during Middle Miocene to Recent. Observed tectonic subsidence calculated from backstripping analysis.
147
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
Serrania de Perija
s a c u L n a S e d i a n a rr e S 0.00
THERMAL MIDDLE MIOCENE QUATERNARY SUBSIDENCE (m)
Maracaibo Basin
e s d A n a r i d e M
Cucuta
0.00
-104.00 0.00
-
0
Bucaramanga
0.00
Barinas Basin
100
200 Km
-172.00
50.00
0.00
0.00 0.00
50.00
47.00
0.00
0.00
Arauca
0.00
0.00
0.00
0.00 0.00
-139.00 149.00
Medellin 0.00
0.00 0.00
124.00 -80.00
0.00
0.00 0.00 26.00
-69.00 -119.00
0.00
0.00
0.00
0
0.00 0.00
-121.00 0.00
0.00
Tunja
-302.00
0.00
-18.00
0.00
0.00 0.00
-291.00
Manizales
-309.00
-281.00
Yopal
0.00 0.00
0.00
0.00
0.00 0.00
0.00
Ibague
0.00
0.00
-289.00
0.00
0.00
-125.00 102.00 -83.00 -109.00
0.00
Bogota
0.00
0
0.00 0.00
0
0.00 109.00
0.00
0.00
0.00
0.00
Villavicencio
900
Llanos Orientales Basin (LLA)
0.00
0.00
00.00
0.00
0.00 0.00
0.00
0.00
0.00 0.00
Neiva
a e n e d r a a i n c a a M r r e a S L
800
800
900
1000
0.00
San Jose del Guaviare
1100
1200
1300
1400
1500
Figure 4.9: Middle Miocene to Recent predicted thermal subsidence after Mesozoic rifting, in meters, calculated from forward modelling using lithosphere-stretching factors discussed in Chapter 2. Negative values mean tectonic rock-uplift.
6.
NEOGENE FLEXURAL SUBSIDENCE
If the EC was generated as a mountain range during Neogene as proposed in literature ( e.g. Cooper et al., 1995), the lithosphere would bend due to the topographic loading (Price, 1973). As a result, depressions would be formed at both sides of a mountain chain (MV and LLA Neogene basins) in which sediments would accumulate. In this section I quantitatively test this hypothesis using a model that assumes the lithosphere behaves as an elastic sheet floating on a fluid asthenosphere (Beaumont, 1981) and the observed Neogene subsidence of the MV and LLA basins. I tested the following scenarios: (1) Observed tectonic subsidence of the MV and LLA basins was produced by present day topographic load of the Central and EC. (2) Observed tectonic subsidence of the MV and LLA basins was only produced by present day topographic load of the EC. (3) Observed tectonic subsidence of the MV and LLA basins was only produced by present day topographic load of the EC and the lithosphere behaved as a broken elastic plate under the EC.
148
Chapter 4
6.1. FLEXURAL MODELS PRODUCED BY THE PRESENT-DAY TOPOGRAPHIC LOAD OF THE EASTERN AND CENTRAL CORDILLERAS Initially I test the hypothesis that the Neogene tectonic subsidence of the MV and LLA basins was produced by lithosphere flexural subsidence due to the present day topographic load of the Eastern and Central cordilleras. 6.1.1. M ethods
I modelled the flexural subsidence along 13 2D sections using the program Cobra developed by Zoetemeijer (1993, 1998). An explanation of the flexural modelling pri nciples and a description of the program is included in Chapter 3. Location of the sections is shown on Figure 4.1. The modelling parameters used are shown in Table 4.1. I applied the present day topography of the Eastern and Central cordilleras along each cross-section to an elastic plate representing the lithosphere. I compared the calculated flexural deflection to the thickness of the Middle Miocene to present-day sedimentary fill of the LLA and MV basins. I changed the effective elastic thickness of the plate until an acceptable fit was obtained between the calculated flexural deflection and the Neogene sedimentary thickness. Initially I tried to use a constant effective elastic thickness, but after several runs the fit was not satisfactory. Then I changed the thickness of the elastic plate as a function of the horizontal distance along the section. After several trials I obtained an acceptable fit. The model assumes the flexural deflection of the elastic plate is instantaneous. Because I applied the present-day topography as a load the model represents the flexural response of the lithosphere at the present time. In continental foreland basins the topographic load alone is usually enough to explain the observed flexural subsidence, but in flexural basins associated with subduction of oceanic lithosphere, subsurface loads, plate boundary forces or moments are usually required (Karner and Watts, 1983; Royden, 1993). The observed subsidence of Neogene basins was fitted assuming surface topographic loads without any subsurface load, boundary force or moment. As an additional constraint to the model the program allows to calculate the gravity anomaly produced by the flexured lithosphere. For that purpose the program applied the algorithm developed by Parker (1972), making use of additional data. These additional data are: (1) The crustal thickness as a function of the horizontal distance along the section; and (2) Densities of the crust, the mantle, the mountain range (topographic load) and the sediments filling the flexural basins created. Gravity data were taken from the gravity map by Kellogg et al. (1991), crustal thickness was taken from a Moho depth map calculated by Salvador (1991) using published gravity and refraction sections in the south of Colombia (Meyer et al., 1973; Ocola et al., 1973; Ramirez and Aldrich, 1973; Meissner et al., 1973, 1976; Mooney et al., 1979; Flueh et al., 1981). Density values applied in the flexural models are displayed in Table 4.1. ASSUMPTIONS -Continuous vs. broken plate model, two possibilities tested: (1) Broken plate with a free boundary located westward on the Romeral suture zone for models assuming a continuous plate under the Eastern Cordillera and (2) Broken plate under the Eastern Cordillera. -Topographic load: two possibilities tested: (1) Eastern and Central cordilleras load, (2) Eastern Cordillera load. -No water depth (Neogene depositional environments were dominantly continental to transitional) except 60 to 80 m paleo-water depth during deposition of shallow marine mudstones of the Leon Fm. (Middle Miocene). -No horizontal intraplate force. -No vertical shear force at the free end of the plate. -No bending moment at the free end of the plate. -Sediment density applied as a load filling the basin, sediment thickness was used to compare to the calculated flexural deflection. -No subsurface loads. Table 4.1: Assumptions and parameters used in lithosphere flexural models.
149
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes PARAMETERS -Length of the plate: 1000 km. -Size of the finite difference grid cell: 5 km. -Densities (kg/m3) Sediments: 2250 Topographic load: 2550 to 2650 Crust: 2770 Mantle: 3300 -Effective elastic thickness and topographic load variable as shown in Figs. 4.10 t o 4.14 a nd Tables 4.2 to 4.4. -Young modulus of the plate: 7x10 10 N/m2 -Poisson ratio of the plate: 0.25 Table 4.1: Continued 6.1.2. Resul ts
Results of these models are presented in Figure 4.10. Table 4.2 shows the effective elastic thickness adopted to obtain the fit shown i n Figure 4.10. In general the fit between the calculated flexural subsidence of the lithosphere and thickness of the Middle Miocene to present-day sediments of the LLA and MV is very good for the eastern LLA basin. However, the model results contain some misfit in the MV area. Comparison of the calculated gravity anomaly with the observed Bouguer anomaly indicates a good correlation at a wavelength of the same order of magnitude as the lithosphere flexural deflection. However the observed gravity represents short wavelength variations that are not comparable to the theoretical gravity values. Probably this high frequency gravity variation is due to shallow lateral changes in crustal density as can be inferred from a surface geological map. The surface geology of the EC is characterized by heterogeneous rocks of different densities. Metamorphic and igneous rocks crop out in the Santander, Floresta, Garzon and Quetame massifs, forming the preMesozoic basement, while less dense sedimentary rocks of Palaeozoic to Cenozoic age constitute most geological units at surface. These sh allow crustal lateral density changes are not considered in the model to calculate the theoretical gra vity anomaly. Also the model includes a calculation of the stress produced by bending the elastic plate (Fig. 4.10). Distance Distance EET EET from from Section (km) Section (km) Romeral Romeral (km) (km) 1 230 15 5 204 5 1 600 15 5 360 5 1 700 50 5 370 25 1 800 50 5 500 25 1 1200 50 5 740 55 2 230 5 5 1200 55 2 390 5 6 173 5 2 400 20 6 370 5 2 500 20 6 390 25 2 550 30 6 520 25 Section number is shown in Fig. 4.1. EET: effective elastic thickness. Table 4.2: Calculated effective elastic thickness (EET) in km.
150
Section 9 9 9 9 9 9 10 10 10 10
Distance from Romeral (km) 231 300 330 400 600 1200 218 390 400 600
EET (km) 5 5 15 15 40 40 5 5 15 15
Chapter 4
Section 2 2 3 3 3 3 3 3 4 4 4 4 4 4
Distance from Romeral (km) 600 1200 172 380 390 525 750 1200 288 480 500 550 850 1200
EET (km)
Section
50 50 5 5 25 25 50 50 5 5 25 25 50 50
6 6 7 7 7 7 7 7 8 8 8 8 8 8
Distance from Romeral (km) 900 1200 240 420 430 650 800 1200 332 450 470 600 900 1200
EET (km) 55 55 5 5 25 25 55 55 5 5 25 25 50 50
Section 10 10 11 11 11 11 11 11 12 12 12 12 12 12 13 13 13 13
Distance from Romeral (km) 700 1200 296 400 410 571 700 1200 239 340 350 509 650 1200 237 480 530 1200
EET (km) 40 40 2 2 15 15 40 40 3 3 15 15 40 40 5 5 40 40
Section number is shown in Fig. 4.1. EET: effective elastic thickness. Table 4.2: Continued
6.2. FLEXURAL MODELS PRODUCED BY THE PRESENT-DAY TOPOGRAPHIC LOAD OF THE EASTERN CORDILLERA ONLY In order to improve the misfit between calculated flexural deflection and thickness of Neogene deposits in the MV, I re-calculated the models including only the topographic load of the EC. This also was helpful to understand the effect of the topographic load represented by the Central Cordillera (on the assumption that the topography of the Central Cordillera remained the same since Palaeogene time, thus assuming no additional load was applied during Neogene time in this region). Results show a good fit between the calculated flexural deflection of the lithosphere and the thickness of the Middle Miocene to present-day sedimentary fill of the LLA and MV basins, significantly reducing the misfit from earlier models (compare fit in the MV in Fig. 4.10 a nd Fig. 4.11). 6.2.1. F lexur al defl ection of t he lith ospher e as a mech a ni sm of tectonic subsidence dur ing N eogene time
The good fit obtained for the model ( Fig. 4.11) indicates that flexural subsidence produced by the present-day topographic load of the E C can explain the observed Neogene subsidence of these basins. A better fit between the calculated flexural deflection and the observed thickness of Neogene sedimentary record support the hypothesis that the topography of the Central Cordillera has not significantly changed since Palaeogene times. Of course, Neogene sediments derived from erosion of the Central Cordillera suggest a reduction of topography that must have been balanced by tectonic rock-uplift without significantly changing the topographic elevation. Almost all the sedimentary cover
151
o(
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes n
SECTION 1
(t
SECTION 2
BOUGUER ANO MALY
n
BOUGUER ANO MALY ref.level :
0
0. m
ref.level :
0
0. m
e -70
-70
-140
-140
ia a 0.
200.
400.
600. 800. distance (km)
1000.
1200.
0.
FLEXURE
)l o
200.
400. 600. 800. distance (km)
1000.
1200.
200.
400.
1000 .
1200.
FLEXURE
5000
5000
ne 0
c d
0
-5000
-5000
-10000 0.
200.
400.
600.
800.
1000 .
1200.
0.
EFFECTIVE ELAS TIC THICKNESS 80 60
60
40
40
ne
20
20
t
0
0 0.
200.
400.
600.
800.
distance (km)
1000 .
0.
1200.
200.
600.
800.
1000 .
1200.
SECTION 4
BOUGUER ANO MALY
BOUGUER ANO MALY ref.level:
0
e
400.
distance (km)
SECTION 3 t
800.
EFFECTIVE ELAS TIC THICKNESS
80
f
600.
0. m
ref.level :
0.
-70
-80.
-140
-160.
0. m
a 0.
200.
400.
600.
800.
distance (km)
1000 .
1200.
0.
800.
1000 .
1200.
FLEXURE
ne
0
0
d
-5000
-5000
0.
200.
400.
600.
800.
1000.
0.
1200.
EFFECTIVE ELAS TIC THICKNESS
e
600.
5000
5000
c
400.
dista nce (km )
FLEXURE
)
200.
200.
400.
600.
800.
1000 .
1200.
1000.
1200.
EFFECTIVE ELAS TIC THICKNESS
80
80
60
60
40
40
20
20
0
0 0.
200.
400.
600.
800.
1000.
1200.
0.
dista nce (km )
200.
400.
600.
800.
distance (km)
Figure 4.10: 2D flexural models produced by the present day topographic load of the Eastern and Central cordilleras. Top panel: Dots represent observed Bouguer anomaly and continuous line calculated Bouger anomaly in mgals. Middle panel: basement deflection and topography in meters. Dots represent the observed thickness of Middle Miocene to R ecent deposits in meters. Bottom panel: effective elastic thickness in km. Location of each section in Figure 4.1.
of the Central Cordillera was removed. The eastward increase in thickness of Neogene sediments of the Middle MV and their onlaping relations on the Central Cordillera suggest that the Middle MV was
152
o
Chapter 4 f f
eastward tilted. Model results ( Fig. 4.11) suggest that the Central Cordillera produced mainly a local effect butn not regional flexural subsidence e ffect. The reduced elastic thickness obtained for the western border of the MV and Central Cordillera explain this more local subsidence and make flexural model results difficult to extrapolate westward in the area of the Central Cordillera. n
SECTION 6
SECTION 5 BOUGUER ANO MALY
e
BOUGUER ANO MALY ref.level:
0.
0. m
ref.leve l :
0.
0. m
e -100.
-80.
-200.
-160.
a a 0.
200.
400.
600.
800.
1000.
0.
1200.
200.
400.
FLEXURE
800.
1000.
1200.
1000 .
1200.
FLEXURE 5000
5000
0
0
d
600.
distance (km)
distance (km)
d -5000
-5000
0.
200.
400.
600.
800.
1000 .
0.
1200.
200.
400.
600.
800.
EFFECTIVE ELAS TIC THICKNESS
EFFECTIVE ELAS TIC THICKNESS 80
80
60
60
40
40
20
20
0 0.
200.
400.
600.
800.
1000 .
0
0.
1200.
200.
SECTION 7
1000 .
1200.
BOUGUER ANO MALY ref.level :
0.
0. m
-100.
-160.
-200. 600.
800.
1000.
ref.leve l :
0.
-80.
1200.
0.
distance (km)
FLEXURE
800.
SECTION 8
BOUGUER ANOMALY
400.
600.
distance (km)
dista nce (km )
200.
400.
0. m
200.
400. 600. 800. distance (km)
1000 .
1200.
200.
400.
800.
1000 .
1200.
800.
1000 .
1200.
FLEXURE
5000
5000
0
0
-5000
-5000
200.
400.
600.
800.
1000.
0.
1200.
600.
EFFECTIVE ELAS TIC THICKNESS
EFFECTIVE ELAS TIC THICKNESS 80
80
60
60
40
40
20
20
0
0 200.
400.
600.
800.
1000.
1200.
distance (km)
0.
200.
400.
600.
distance (km)
Figure 4.10: Continued
153
f f
n ( r
t t Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes ( r
e (
en (
t
SECTION 5
SECTION 6
t
BOUGUER ANO MALY 0.
e
BOUGUER ANO MALY
) ref.level:
0. m
e y
-100.
i
ref.leve l :
0. -80.
i -200.
-160.
a a 0.
m l
200.
400.
600.
800.
1000.
1200.
l
n
200.
400.
600.
800.
1000 .
0.
1200.
f
600.
800.
60 40
n e
20
t
20
0 0.
200.
400.
600.
800.
1000 .
0
0.
1200.
200.
SECTION 7 t
400.
600.
800.
distance (km)
dista nce (km )
1000 .
1200.
SECTION 8 t
BOUGUER ANOMALY
BOUGUER ANO MALY ref.level :
0.
0. m
e
-80.
ref.leve l :
0.
0. m
-100.
a
-160. 200.
400.
600.
800.
1000.
-200.
1200.
0.
distance (km)
FLEXURE
200.
400. 600. 800. distance (km)
1000 .
1200.
200.
400.
800.
1000 .
1200.
800.
1000 .
1200.
FLEXURE
5000
5000
)
) ne
0
c
0
c d
-5000
200.
400.
600.
800.
1000.
-5000
0.
1200.
600.
EFFECTIVE ELAS TIC THICKNESS
EFFECTIVE ELAS TIC THICKNESS 80
80
)
60
60
40
e
400.
EFFECTIVE ELAS TIC THICKNESS
40
e
200.
80
80 60
)
1200.
d -5000
-5000
f
d
1000 .
0
EFFECTIVE ELAS TIC THICKNESS
ne
1200.
c
0.
a
1000.
5000
0
d
e
600. 800. distance (km)
e
n
t
400.
)
e c
200.
FLEXURE
o
5000
)
0.
m
distance (km)
FLEXURE
o
n
0. m
40
20
e
0 200.
400.
600.
800.
1000.
20 0
1200.
distance (km)
0.
200.
400.
600.
distance (km)
Figure 4.10: Continued
154
Chapter 4 SECTION 13 t
BOUGUER ANOM ALY ref.level :
0.
e
0. m
-160. -320
a
0.
200.
400.
600.
800.
1000 .
1200.
400.
600.
800.
1000 .
1200.
1000 .
1200.
FLEXURE 5000
) e n c
0
-5000
d -10000 0.
200.
EFFECTIVE ELAS TIC THICKNESS 80 60 40
e
20 0 0.
200.
400.
600.
800.
distance (km)
Figure 4.10: Continued 6.2.2. Calcu lated bendi ng str ess
The program also calculated the bending stresses of the assumed thin elastic plate ( Fig. 4.11). In figures showing results of flexural models, the fiber stress of the modelled profiles rel ative to the distance from the plate edge, includes indications for the onset of yielding at σ* = 1Gpa. For most of the profiles the bending stress generated in the flexed thin elastic plate is close or even grea ter than the normal rock strength in the place where the EC reaches its maximum topographic elevation ( Fig. 4.11). Probably this result suggests that a model with a broken plate is more appropriate to represent the flexural deflection produced by the present-day topography of the EC. Additionally except for the most northern profiles the calculated deflection in the Middle MV is a little in excess of the observed subsidence. If the lithosphere in the western part of the model were thinner than in the eastern part, deflection would have given a better fit. This suggests that probably an initial “subduction” of the LLA lithosphere is occurring below the EC and the Middle MV. From structural analysis of a regional balanced cross-section and comparison with analogue model experiments, Colleta et al. (1990) have suggested a possible “continental subduction” under the EC. These authors did not reach conclusions about the direction of such a possible subduction. Flexural modelling results support this hypothesis. However it is difficult to constrain the subduction direction. I also tested with modelling the hypothesis of a broken plate under the EC.
6.3. FLEXURAL SUBSIDENCE MODELS WITH A BROKEN PLATE UNDER THE EASTERN CORDILLERA 6.3.1. M ethods
The possibility of a broken plate under the EC as discussed above was also tested through flexural models assuming a broken plate with a free end under the Cordillera, using an additional feature of the Program Cobra not mentioned before. This is the possibility to model the lithosphere as a thin elastic broken plate to simulate plate-subduction (Zoetemeijer, 1993). The program allows applying
155
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
forces or bending moments to the free end of the thin elastic broken plate model. However, I did not apply forces or bending moments at the end of the plate. SECTION 1
) l BOUGUER ANOMALY a g m 0. ( y -70. l a m -140. o n a 500.
ref.level :
1000.
0. m
) l a g m ( y l a m o n a
SECTION 2 BOUGUER ANOMALY 0. ref.level :
-70. -140. 0.
500.
distance (km)
) 5000. m ( n o 0. i t c e l f -5000. e d
0. m
1000.
1500.
distance (km)
FLEXURE
) m ( n o i t c e l f e d
FLEXURE 5000. 0.
-5000.
-10000. 500.
1000.
0.
500.
distance (km)
) m k ( t e e
80.
) m k ( t e e
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0. 0.
1000.
500.
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1. 500.
1000.
1500.
distance (km)
distance (km)
) a P G ( s s e r t s
1500.
distance (km)
EFFECTIVE ELASTIC THICKNESS
500.
1000.
1000.
FIBER STRESS 1. 0. -1. 0.
distance (km)
500.
1000.
1500.
distance (km)
Figure 4.11: 2D flexural models produced by the present day topographic load of the Eastern Cordillera only. Top panel: Dots represent observed Bouguer anomaly and continuous line calculated Bouguer anomaly in mgals. Second panel: basement deflection and topography in meters. Dots represent the observed thickness of Middle Miocene to Recent deposits in meters. Third panel: effective ela stic thickness in km. Bottom panel: Calculated bending (fiber) stress in Gpa. Location of each section in Figure 4.1. 6.3.2. Resul ts
Results of these models are presented in Figure 4.12. The fit obtained for the eastern side of the model was excellent and very similar to the previous models in which the elastic plate under the EC was continuous. For the western part of the model, except for the most northern two sections, the fit between the flexural deflection and the thick ness of Neogene sediments of the MV is also very good and better than previous models (compar e Figs. 4.11 and 4.12). This supports the hypothesis of “mantle lithosphere subduction” under the EC.
156
Chapter 4
SECTION 3
) l BOUGUER ANOMALY a g m 0. ( y -70. l a m -140. o n a 0. 500.
ref.level :
1000.
0. m
) l a g m ( y l a m o n a
SECTION 4 BOUGUER ANOMALY 0. ref.level :
-80. -160. 500.
) m ( n o i t c e l f e d
FLEXURE
0.
500.
) m k ( t e e
5000.
0.
-5000.
500.
1000.
) m k ( t e e
60. 40. 20. 0. 500.
80. 60. 40. 20. 0.
500.
1000.
) a P G ( s s e r t s
1. 0. -1. 500.
1000.
1500.
distance (km)
FIBER STRESS
0.
1500.
EFFECTIVE ELASTIC THICKNESS
distance (km)
) a P G ( s s e r t s
1000. distance (km)
EFFECTIVE ELASTIC THICKNESS
0.
1500.
FLEXURE
distance (km)
80.
1000. distance (km)
distance (km)
) 5000. m ( n o i t 0. c e l f e d-5000.
0. m
FIBER STRESS 1. 0. -1.
1000.
500.
1000.
1500.
distance (km)
distance (km)
Figure 4.11: Continued
6.4. LATERAL DISTRIBUTION OF EFFECTIVE ELASTIC THICKNESS AT PRESENT The estimated eff ective elastic thickness, reaching a maximum value of 50-55 km in the eastern LLA basin is shown on Figure 4.13. In comparison the effective elastic thickness in the EC shows values of 25 km or less. In the area of the Central Cordillera results indicate low elastic thickness of 5 km. Values of estimated effective elastic thickness of less than 10 km occur in the northern and southern parts of the EC, where major strike-slip faults such as the Santa Marta-Bucaramanga, Bocono and Altamira faults are located. These results will be interpreted in terms of lithosphere strength.
157
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
SECTION 5 ) l a g m ( y l a m o n a
SECTION 6
BOUGUER ANOMALY 0. ref.level :
0. m
-90. -180. 500.
1000.
) l a g m ( y l a m o n a
BOUGUER ANOMALY 0. ref.level :
-160. 0.
5000.
FLEXURE
) m ( n o i t c e l f e d
0.
-5000.
500.
80.
0.
-5000.
0.
1000.
) m k ( t e e
60. 40. 20. 0.
80. 60. 40. 20. 0.
0.
1000.
500.
1000.
distance (km)
FIBER STRESS
) a P G ( s s e r t s
1. 0. -1. 500.
1000.
EFFECTIVE ELASTIC THICKNESS
distance (km)
) a P G ( s s e r t s
500. distance (km)
EFFECTIVE ELASTIC THICKNESS
500.
1000.
FLEXURE 5000.
distance (km)
) m k ( t e e
500. distance (km)
distance (km)
) m ( n o i t c e l f e d
0. m
-80.
1000.
FIBER STRESS 1. 0. -1.
distance (km)
0.
500.
1000. distance (km)
Figure 4.11: Continued
7.
FLEXURAL SUBSIDENCE PRODUCED BY GRADUAL SURFACE-UPLIFT OF THE EASTERN CORDILLERA DURING NEOGENE TIME
Foreland basins often contain a valuable record of lithosphere processes in their sedimentary sequences. Especially because of the continual flux of sediments, foreland basins are excellent recorders of changes in geodynamic processes, such as changes in lithosphere strength and mountain building processes (Price, 1973; Beaumont, 1981). The geometry of the internal stratigraphy of the Neogene sedimentary fill during several time intervals recorded the progressive lithospheric flexural deflection as a response to the progressive growth of the topographic load during surface-uplift of the EC. I modelled the evolution of lithosphere flexure due to a growing topography of the EC along a 2D regional section using several stratigraphic markers interpreted from a regional seismic traverse of the
158
Chapter 4
) l a g m ( y l a m o n a
SECTION 7
) l a g m ( y l a m o n a
BOUGUER ANOMALY 0. ref.level :
0. m
-80. -160. 0.
500.
1000.
1500.
SECTION 8 BOUGUER ANOMALY 0. -70.
200.
400.
-5000.
500.
1000.
1500.
200.
400.
distance (km)
) m k ( t e e
80.
1200.
600.
1000.
1200.
800.
distance (km)
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
0.
500.
1000.
1500.
200.
400.
distance (km)
) a P G ( s s e r t s
1000.
800.
FLEXURE ) 5000. m ( n o i t 0. c e l f e d-5000.
0.
0.
600.
distance (km)
FLEXURE
5000.
0. m
-140.
distance (km)
) m ( n o i t c e l f e d
ref.level :
) a P G ( s s e r t s
1. 0. -1. 500.
1000.
800.
1000.
1200.
1000.
1200.
distance (km)
FIBER STRESS
0.
600.
FIBER STRESS 1. 0. -1.
1500.
200.
distance (km)
400.
600.
800.
distance (km)
Figure 4.11: Continued
LLA calibrated with well data (Fig. 4.6). The evolution of the growing topography and lithosphere strength through Neogene time c an be outlined from this model.
7.1 DATA The section is located approximately close to section 6 ( Fig 4.1). Fig 4.6 s hows a line drawing interpretation from the regional seismic traverse crossing the LLA basin. I identified the Leon Fm and five stratigraphic intervals within the Guayabo Fm. I measured the two-way t ravel time for each of the marker reflectors and using depth information for several wells on the section ( Table 4.3), I calculated depth of each stratigraphic marker for several points along the section. Taking into account that the Leon Fm was deposited under shallow marine conditions I applied also a water depth load as indicated in Table 4.4. For the continental facies of the Guayabo Fm I did not apply any water load.
159
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes ) l a g m ( y l a m o n a
SECTION 9
SECTION 10
BOUGUER ANOMALY 0. ref.level :
-80.
0. m
-160. 200.
400.
600.
800.
1000.
) l a g m ( y l a m o n a
BOUGUER ANOMALY 0. ref.level :
-90. -180. 200.
400.
distance (km)
) m ( n o i t c e l f e d
5000.
) m ( n o i t c e l f e d
0.
-5000.
400.
600.
800.
5000.
1000.
0. -5000. -10000.
1000.
200.
400.
600.
800.
1000.
distance (km)
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
80. 60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
200.
) a P G ( s s e r t s
800.
FLEXURE
distance (km)
) m k ( t e e
600.
distance (km)
FLEXURE
200.
0. m
400.
600.
800.
1000.
200.
400.
distance (km)
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1. 200.
400.
600.
800.
600.
800.
1000.
distance (km) FIBER STRESS 1. 0. -1.
1000.
200.
distance (km)
400.
600.
800.
1000.
distance (km)
Figure 4.11: Continued
Unfortunately the internal stratigraphy of the Neogene sedimentary fill has not been dated and age determinations of stratigraphic markers interpreted from the seismic are absent. On the assumption of a constant sedimentation rate during Neogene, I estimated the age of these stratigraphic markers in order to try to compare them with the surface-uplift history of the Sabana de Bogotá area. Such a history has been recorded by the Neogene sedimentary and palynological record ( e.g. Helmens, 1990, Andriessen et al., 1993).
7.2. METHOD For each of the stratigraphic markers I modelled the flexural deflection of the lithosphere necessary to fit the thickness accumulated between the bottom of the Leon Fm and each one of the selected stratigraphic markers. I applied the same principles and program as the previous modelling. I systematically changed the topography and the effective elastic thickness until an acceptable fit was obtained between the calculated flexural deflection and the thickness of each stratigraphic interval.
160
Chapter 4
) l a g m ( y l a m o n a ) m ( n o i t c e l f e d
SECTION 11 BOUGUER ANOMALY 0. ref.level :
-90.
0. m
-180. 200. 5000.
400.
600.
800.
SECTION 12 BOUGUER ANOMALY 0.
) m ( n o i t c e l f e d
0. -5000.
ref.level :
-80.
0. m
-160. 200.
1000.
distance (km)
FLEXURE
) l a g m ( y l a m o n a
5000.
400.
600.
800.
1000.
distance (km)
FLEXURE
0.
-5000.
-10000. 200.
400.
600.
800.
200.
1000.
400.
80.
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20.
1000.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
0. 200.
400.
600.
800.
) a P G ( s s e r t s
1. 0. -1. 400.
600.
800.
400.
600.
800.
1000.
distance (km)
FIBER STRESS
200.
200.
1000.
distance (km)
) a P G ( s s e r t s
800.
distance (km)
distance (km)
) m k ( t e e
600.
FIBER STRESS 1. 0. -1. 200.
1000.
400.
600.
800.
1000.
distance (km)
distance (km)
Figure 4.11: Continued
7.3. RESULTS Figure 4.14 sho ws the modelled results and Table 4.4 co ntains the effective elastic thickness and topography obtained for the best fit of each model. In general the fit between calculated flexural deflection and thickness of sediments filling the LLA basin is good. A comparison of the model results for different stratigraphic markers allows to outline the evolution of the growing topography of the EC through Neogene time, and the lithosphere strength evolution revealed by temporal changes of the effective elastic thickness. I tried to correlate such a history to the rock-uplift history of the EC as revealed by the stratigraphic and palynological record of the Sabana de Bogotá area. In order to do such a correlation I estimated ages for the stratigraphic markers of the Guayabo Fm assuming constant sedimentation rates.
161
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
SECTION 13 ) l a g m ( y l a m o n a
BOUGUER ANOMALY 0. ref.level :
0. m
-100. -200. 200.
400.
600.
800.
distance (km)
) 5000. m ( n 0. o i t c -5000. e l f e d
FLEXURE
-10000. 200.
400.
600.
800.
distance (km)
) m k ( t e e
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0. 200.
400.
600.
800.
distance (km)
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1. 200.
400.
600.
800.
distance (km)
Figure 4.11: Continued
8. UPLIFT EVOLUTION FROM FLEXURAL MODELING, FISSION TRACK AND GEOLOGICAL DATA 8.1. SURFACE-UPLIFT EVOLUTION OF THE EASTERN CORDILLERA AS INFERRED FROM FLEXURAL MODELLING In the LLA and MV, Neogene flexural subsidence created accommodation space for sediment accumulation. Probably the maximum topographic elevation in those basins occurs at present time close to 200 m above sea-level. At the end of deposition of the Leon F m (end of Middle Miocene; Cooper et al., 1995) the model predicts 500 m of elevation in the area of t he EC (Fig. 4.14). Later at the time of deposition of the str atigraphic marker 1 of the lower part of the Guayabo Fm (Late Miocene?), the model predicts a surface-uplift of 2000 m in the western flank of the former extensional basin and 2500 m in its eastern flan k (Fig. 4.14). In the middle axial Tunja region of the
162
Chapter 4
EC, where the Palaeogene sedimentary record has been partially preserved, I assumed a topography similar to that interpreted for the palynological record of the Late Mioc ene Marichuela Fm (Helmens, 1990; Andriessen et al., 1993), with an elevation not greater than 500 m (Fig. 4.14).
Well
La María-1 La Cabaña-1
La Gloria-1
Centauro-1
La Punta-1
Guarimena-1
Morichito-1
Caño Barulia-1
Dist Reflector from Top León Fm. Guayabo 1 Romeral (km) (sec) (m) (sec) (m) 216.40 221.40 226.40 229.50 233.75 235.85 240.85 243.35 248.35 253.35 254.20 259.20 264.20 269.20 274.20 277.80 279.40 284.40 289.40 294.40 299.40 301.05 306.05 311.05 311.85 315.55 328.55 333.55 338.70 343.70 348.70 353.70 355.90
2.00 1.97 1.84 1.84 1.80 1.78 1.65 1.62 1.60 1.54 1.52 1.49 1.48 1.44 1.42 1.39 1.35 1.31 1.29 1.28 1.23 1.23 1.24 1.22 1.21 1.19 1.07 1.02 0.99 0.92 0.90 0.86 0.86
4088.86 3933.70 3313.40 3453.38 2926.08 3054.39 2548.35 2441.77 2372.75 2175.19 2112.43 2176.27 1991.39 1876.19 1820.71 1740.08 1706.88 1539.70 1492.80 1469.81 1359.38 1319.78 1380.88 1338.18 1150.62 1276.27 1052.84 970.33 923.57 821.95 853.44 742.58 742.58
1.46 1.44 1.38 1.38 1.33 1.31 1.31 1.29 1.29 1.28 1.27 1.23 1.21 1.18 1.14 1.13 1.11 1.09 1.06 1.05 1.04 1.03 1.03 1.03 1.03 1.03 0.95 0.91 0.87 0.81 0.80 0.78 0.78
1933.07 1876.19 1713.88 1713.88 1587.85 1539.70 1539.70 1492.80 1492.80 1469.81 1447.13 1359.38 1317.26 1256.20 1178.61 1159.88 1123.19 1087.52 1035.87 1019.13 1002.63 986.36 986.36 986.36 986.36 986.36 864.26 808.25 755.35 681.41 669.67 646.69 646.69
Reflector Guayabo 2 (sec)
(m)
1.07 1.06 1.02 1.03 1.03 1.01 0.97 0.97 0.97 0.98 0.98 0.98 0.96 0.93 0.92 0.91 0.90 0.90 0.88 0.86 0.85 0.84 0.83 0.82 0.81 0.82 0.79 0.74 0.70 0.65 0.64 0.62 0.62
1052.84 1035.87 970.33 986.36 986.36 954.52 893.49 893.49 893.49 908.42 908.42 908.42 878.77 835.85 821.95 808.25 794.74 794.74 768.29 742.58 730.00 717.60 705.36 693.30 681.41 693.30 658.11 602.58 560.76 511.42 501.92 483.25 483.25
Reflector Guayabo 3
Reflector Guayabo 4
(sec) (m)
(sec)
(m)
0.73 0.70 0.68 0.68 0.70 0.69 0.69 0.68 0.67 0.68 0.68 0.68 0.68 0.67 0.63 0.62 0.62 0.62 0.62 0.61 0.60 0.59 0.58 0.57 0.55 0.53 0.50 0.49 0.44 0.39 0.38 0.35 0.35
0.42 0.41 0.39 0.40 0.41 0.41 0.40 0.39 0.39 0.39 0.39 0.40 0.39 0.39 0.40 0.38 0.37 0.35 0.34 0.32 0.32 0.32 0.31 0.31 0.30 0.31 0.29 0.27 0.23 0.19 0.18 0.17 0.17
316.31 308.67 293.52 301.07 308.67 308.67 301.07 293.52 293.52 293.52 293.52 301.07 293.52 293.52 301.07 286.00 278.52 263.66 256.27 241.56 241.56 241.56 234.24 234.24 226.93 234.24 219.63 205.06 175.94 146.68 139.33 131.95 131.95
591.91 560.76 540.65 540.65 560.76 550.64 545.63 540.65 530.79 540.65 540.65 540.65 540.65 530.79 492.53 483.25 483.25 483.25 483.25 474.07 465.01 456.04 447.17 438.40 421.14 404.22 379.44 371.33 331.75 293.52 286.00 263.66 263.66
Table 4.3. Two way travel time and depth of the reflectors selected from seismic lines along section B. Guayabo 5 reflector is not included since represent the present day earth surface
163
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
Distance from Romeral Region / sub basin (km) Middle Magdalena Valley " " " "
Top Fm León (end Middle Miocene)
Guayabo 1 Guayabo 2 Guayabo 3 Guayabo 4 Guayabo 5 (Late (Early (Middle (Late (present Miocene?) Pliocene?) Pliocene?) Pliocene?) day)
top EET pwd top EET top EET top EET top EET top EET (m) (km) (m) (m) (km) (m) (km) (m) (km) (m) (km) (m) (km)
397
0
10
0
0
15
0
17
0
23
0
24
pdt
25
397.2
0
10
0
0
15
0
17
0
20
0
24
"
25
" " W flank Eastern Cordillera " " "
420
0
10
0
0
15
0
17
0
int
0
24
"
25
452.5
int
10
int
int
15
1500
17
2000
18
2000
24
"
25
460
int
10
int
int
15
int
17
int
18
1500
24
"
25
"
462.5
500
10
int
2000
15
int
17
int
18
1500
24
"
25
" " " Tunja Axial region " " "
466
500
10
int
int
15
int
17
int
18
1500
24
"
25
515
500
10
int
500
15
500
17
1300
18
2200
24
"
25
520
500
10
int
int
15
500
17
int
18
2200
24
"
25
"
"
530
500
10
int
int
int
500
17
1300 int
2200
24
"
int
" " E flank Eastern Cordillera " " "
550
500
10
int
int
int
3000
17
3300 int
3000
24
"
int
568.5
500
10
int
int
int 3000
17
3300 int
3000
24
"
int
588.5
500
10
80
3200
17
3300 int
int
24
"
int
590
int
10
int
int
int
Int
17
3300 int
3300
24
"
int
610
0
10
int
0
int
0
17
0
int
0
24
"
int
int
0
int
0
17
0
int
0
24
"
int
"
"
"
"
"
"
"
Llanos basin
2500 int
"
"
"
615.5
0
"
"
"
620
0
50
int
0
int
0
17
0
int
0
24
"
int
"
"
"
650
0
50
60
0
int
0
17
0
17
0
24
"
int
"
"
"
660
0
50
int
0
int
0
int
0
20
0
24
"
int
"
"
"
669
0
50
int
0
int
0
int
0
int
0
50
"
int
"
"
"
670
0
50
int
0
int
0
int
0
50
0
50
"
int
"
"
"
700
0
50
0
0
30
0
int
0
50
0
50
"
int
"
"
"
730
0
50
0
0
int
0
50
0
50
0
50
"
int
"
"
"
900
0
50
0
0
50
0
50
0
50
0
50
"
55
" " " 1200 0 50 0 0 50 0 50 0 50 0 50 " 55 EET: effective elastic thickness; top: palaeotopograhy; pwd: palaeowater depth, for the other stratigraphic markers palaeowater depth is zero; pdt: present day topography; int: linearly interpolated value depending on the distance from the Romeral suture Table 4.4: Evolution of the topography of the Eastern Cordillera along a regional section according to the 2D flexural models for different reflectors in the Llanos Basinsrepresenting different times during Neogene.
164
Chapter 4
) l a g m ( y l a m o n a
SECTION 2 BOUGUER ANOMALY 0.
ref.level :
0. m
-120. -240. 300.
400.
500.
600.
) l a g m ( y l a m o n a
SECTION 3 BOUGUER ANOMALY 0.
ref.level :
-80. -160. 250.
300.
350.
distance (km)
) m ( n o i t c e l f e d
0. m
400.
450.
500.
550.
500.
550.
500.
550.
500.
550.
distance (km)
FLEXURE
) m ( n o i t c e l f e d
5000. 0. -5000. -10000.
FLEXURE
5000. 0. -5000.
-10000. 300.
400.
500.
600.
250.
300.
350.
distance (km)
) m k ( t e e
80.
450.
distance (km)
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
300.
400.
500.
600.
250.
300.
350.
distance (km)
) a P G ( s s e r t s
400.
) a P G ( s s e r t s
1. 0. -1. 400.
450.
distance (km)
FIBER STRESS
300.
400.
500.
600.
distance (km)
FIBER STRESS 1. 0. -1. 250.
300.
350.
400.
450.
distance (km)
Figure 4.12: 2D flexural models of the western part of the Eastern Cordillera assuming a broken plate under the Eastern Cordillera. Top panel: Dots represent observed Bouguer anomaly and continuous line calculated Bouguer anomaly in mgals. Second panel: basement deflection and topography in meters. Dots represent the observed thickness of Middle Miocene to Recent deposits in meters. Third panel: effective elasti c thickness in km. Bottom panel: Calculated bending (fiber) stress in Gpa. Location of each section in Figure 4.1.
Calculated bending stress started to exceed normal rock strength since the time of deposition of this stratigraphic marker (Fig. 4.14). If some subducting lithosphere exists under the EC probably, it began to subduct during deposition of str atigraphic marker 1. Later during deposition of the stratigraphic marker 2, between the lower and middle parts of the Guayabo Fm a palaeo-topography of maximum 1500 m and 3200 m respectively in the western and eastern flanks of the palaeo-EC fits the observed subsiden ce (Fig. 4.14). In the middle axial Tunja region the model requires only 500 m of palaeo-topography ( Fig. 4.14) as interpreted by reference to the elevation of the Sabana de Bogotá area during deposition of the Tequend ama member of the Tilata Fm during Early Pliocene (Helmens, 1990; Andriessen et al., 1993). At the time of deposition of the stratigraphic marke r 3, between the middle and upper part of the Guayabo Fm, the model requires a palaeo-topography varying between 1300 and 2000 m in the western flank and 3300 m in the eastern flank of the palaeo- EC (Fig. 4.14). In the axial Tunja region a
165
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes ) l a g m ( y l a m o n a
SECTION 4 BOUGUER ANOMALY ref.level :
0.
0. m
-110. -220. 350.
400.
450.
500.
550.
600.
650.
) l a g m ( y l a m o n a
SECTION 5 BOUGUER ANOMALY 0.
ref.level :
-120. -240. 300.
350.
distance (km)
) m ( n o i t c e l f e d
FLEXURE
) 5000. m ( 0. n o i t c -5000. e l f e d -10000.
0. -5000. -10000. 400.
450.
500.
550.
80.
600.
650.
300.
500.
350.
400.
450.
500.
distance (km)
) m k ( t e e
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
350.
400.
450.
500.
550.
600.
650.
300.
350.
distance (km)
) a P G ( s s e r t s
450.
FLEXURE
distance (km)
) m k ( t e e
400.
distance (km)
5000.
350.
0. m
) a P G ( s s e r t s
1. 0. -1. 400.
450.
500.
550.
450.
500.
distance (km)
FIBER STRESS
350.
400.
600.
FIBER STRESS 1. 0. -1.
650.
distance (km)
300.
350.
400.
450.
500.
distance (km)
Figure 4.12: Continued
palaeo-topography of 1300 m fits the observed subsidence (Fig. 4.14). This palaeo-elevation is similar to the palaeo-elevation between 1000 and 1500 m interpreted from the palynological record of the early Late Pliocene Tibagota member of the Tilata Fm in the Sabana de Bogotá area (Helmens, 1990; Andriessen et al., 1993). Model results calculated at the time of deposition of stratigraphic marker 4, in the upper part of the Guayabo Fm suggest 1500 to 2200 m of topography in the western flank of the Cordiller a, 2200 to 3300 m in the eastern flank, and 2200 m of topography in the axial Tunja region ( Fig. 4.14). 2200 m of palaeo-elevation also was interpreted from the palynological record of the Late Pliocene Guasca Member of the upper Tilata Fm in the Sabana de Bogotá region (Helmens, 1990; Andriessen et al., 1993). Finally the present day topography along the section with elevations bet ween 1000 to 3000 m in the western and eastern flanks of the EC and between 2500 to 3 000 in the Tun ja axial region fits the Neogene thickness up to the present day surface of the LLA basin (Fig. 4.14). Probably evidence of small rock-uplift and erosion in some places of the MV and LLA basin ( e.g. Fig. 4.7), evidenced by fluvial terraces, are related to regional isostatic flexural rebound produced by recent erosion of the mountain range.
166
Chapter 4
) l a g m ( y l a m o n a ) m ( n o i t c e l f e d
SECTI0N 6 BOUGUER ANOMALY 0.
ref.level :
0. m
-100. -200. 250.
300.
350.
400.
450.
500.
550.
) l a g m ( y l a m o n a
SECTION 7 BOUGUER ANOMALY 0.
ref.level :
-90. -180. 350.
400.
distance (km)
0. -5000.
550.
500.
550.
500.
550.
500.
550.
-10000. 250.
300.
350.
400.
450.
500.
550.
350.
400.
distance (km)
450. distance (km)
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
250.
300.
350.
400.
450.
500.
550.
350.
400.
distance (km)
) a P G ( s s e r t s
500.
) FLEXURE 5000. m ( n o 0. i t c e l f -5000. e d
FLEXURE 5000.
80.
450. distance (km)
-10000.
) m k ( t e e
0. m
distance (km)
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1. 250.
300.
350.
400.
450.
450.
500.
FIBER STRESS 1. 0. -1.
550.
distance (km)
350.
400.
450. distance (km)
Figure 4.12: Continued
8.2. COMPARISON OF THE UPLIFT AND EXHUMATION EVOLUTION INFERRED FROM FISSION TRACK AND GEOLOGICAL DATA In Chapter 3, fission track data and other evidence of deformation/uplift-exhumation have been discussed. According to these data, during Neogene time pronounced exhumation occurred at all localities of the northern Andes. In the Mérida Andes an exhumation of approximately 2300 m ocurred during the Pliocene (Kohn et al., 1984; Shagam et al., 1984). In the Sabana de Bogotá (axial zone of the EC) the fluvial-lacustrine sediment record registers major tectonic surface-uplift for the period between 5 and 3 Ma (Helmens, 1990; Andriessen et al., 1993). In the Garzón Massif, Van der Wiel (1991) has shown that between 12 Ma and the present time the massif was exhumated approximately 6500 m. After the first Neogene (12 Ma) exhumation pulse a second pulse occurred 6.4 Ma ago. The rate of very fast exhumation must have exceeded the rate of thermal diffusion, producing thermal updoming of the isotherms, and the ages should be interpreted as cooling following exhumation (Shagam et al., 1984; Van der Wiel, 1991). The most complete data from these three localities indicate rapid regional rock-uplift, which agrees with the model results.
167
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes ) l a g m ( y l a m o n a
SECTION 8 BOUGUER ANOMALY 0.
ref.level :
0. m
-110. -220. 400.
450.
500.
550.
) l a g m ( y l a m o n a
SECTION 9 BOUGUER ANOMALY 0.
ref.level :
-110. -220. 250.
300.
distance (km)
) 5000. m ( n 0. o i t c -5000. e l f e d-10000.
80.
FLEXURE
450.
500.
550.
250.
300.
350.
400.
distance (km)
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20. 0.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
400.
) a P G ( s s e r t s
400.
) 5000. m ( 0. n o i t c -5000. e l f e d -10000. distance (km)
) m k ( t e e
350. distance (km)
FLEXURE
400.
0. m
450.
500.
550.
distance (km)
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1. 400.
450.
500.
250.
550.
350.
400.
distance (km) FIBER STRESS 1. 0. -1. 250.
distance (km)
300.
300.
350.
400.
distance (km)
Figure 4.12: Continued
Fission track ages and other geologic evidence indicate initiation of Andean rock-uplift and exhumation of blocks at the end of Cretaceous-Paleocene with local phases of rock-uplift during Eocene to Miocene time, terminating with regional rock-uplift and exhumation in all blocks in the Pliocene-Pleistocene (Shagam et al., 1984). These authors interpreted this to result from uncoupled crustal blocks being uplifted in response to local stress until regional compression led to interlocking of the blocks and their simultaneous rock-uplift during the Pliocene-Pleistocene time. Model results ( Fig. 4.14) agree with these observations. Probably during the Neogene the normal fault systems that delimited the Mesozoic extensional basin were completely inverted in a compressional/transpressional stress field. The most uplifted areas were close to these fault systems. The eastern flank of the Cordillera illustrates these processes. The Chámeza thrust fault system, which was one the eastern extensional Mesozoic master faults limiting the basin, started an episode of total inversion 25 Ma ago (Hossack et al., 1999). Fission track data reported by Hossack et al. (1999) indicates that the sedimentary fill of the former extensional basin close to the fault began to uplift and exhumate at that time. Younger apatite data (15 to 3 Ma) from the more internal parts of the Cordillera suggest later exhumation (Hossack et al., 1999). Finally the surface of the central axial Sabana de Bogotá region was uplifted during Pliocene time (5 to 3 Ma; Andriessen et al., 1993). However, better data are needed to refine models and unravel more precisely the Neogene uplift history.
168
Chapter 4
) l a g m ( 0. y l a-120. m o-240. n a
SECTION 10
SECTION 11
BOUGUER ANOMALY ref.level :
300.
350.
400.
450.
0. m
500.
) l a g m ( y l a m o n a
BOUGUER ANOMALY 0.
ref.level :
-120. -240. 350.
distance (km)
) 5000. m ( 0. n o i t c -5000. e l f e d -10000. 350.
400.
450.
80.
500.
350.
) m k ( t e e
60. 40. 20. 0.
) a P G ( s s e r t s
350.
400.
450.
80. 60. 40. 20. 0.
500.
350.
1. 0. -1. 400.
450.
400.
450.
distance (km)
FIBER STRESS
350.
450.
EFFECTIVE ELASTIC THICKNESS
distance (km)
300.
400. distance (km)
EFFECTIVE ELASTIC THICKNESS
300.
450.
FLEXURE
distance (km)
) m k ( t e e
400. distance (km)
FLEXURE ) m 5000. ( n o 0. i t c e l f -5000. e d -10000.
300.
0. m
500.
) a P G ( s s e r t s
FIBER STRESS 1. 0. -1.
distance (km)
350.
400.
450.
distance (km)
Figure 4.12: Continued
In the Middle MV Middle and Upper Miocene, fluvial sediments were depo sited by a northwarddirected fluvial system, which was fed by the Eastern and Central cordillera s (Figs. 1.1 an d 4.4). Sedimentation was contemporaneous with Central Cordillera volcanism between 10.92 and 6.2 Ma (Gómez et al., 1999). However, the dominantly eastward palaeocurrent directions measured in the Palaeogene (La Paz Esmeraldas, Mugrosa and Colorado Fms of the Middle MV, e.g. Rubiano, 1998; Gualanday Gp of the Upper MV, e.g. Anderson, 1970, 1972) and Neogene (Real Fm of the Middle MV, Rubiano, 1998; Honda, Neiva and Gigante Fms of the Upper MV, e.g. Wellman, 1970; Howe, 1974; Van der Wiel, 1991) sedimentary record of the MV seems to be in contradiction with the hypothesis of gradual surface-uplift of the EC. Van der Wiel (1991) found eastward palaeocurrent directions in the Neogene Honda Fm of the Upper M V and fission track data indicating simultaneous rock-uplift and exhumation of the EC Garzon Massif ( Fig. 2.1). She proposed the following possible explanations for this apparent contradiction: (1) The Garzón Massif is an eastward tilted block, with most of the present-day drainage flow eastward toward the continent interior and only minor amount of sediment flow toward the western side in the Upper MV; (2) Within the EC Garzon Massif there are small sedimentary basins that trapped sediments inhibiting their westward distribution in the Upper
169
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
MV; (3) The volume of sediments and volcanic products from the Central Cordillera was greater than the volume of sediments from the rising EC. The MV surface was tilted toward the east; the ProtoMagdalena River flowed close to the eastern border of the basin capturing all the eastward flowing drainage systems. Westward drainage was poorly developed and possibly only occurred in the easternmost part of the MV in areas of poor exposure (Van der Wiel, 1991). P robably these explanations may also be extended to the whole MV. The axial Sabana de Bogotá region (Figs. 2.1 an d 4.4) was a sedimentary basin until recent time (Helmens, 1990; Andriesen et al. 1 993) trapping sediments and preventing their distribution on the MV. The eastward thickening of the sedimentary record for several Tertiary time intervals in the MV indicates a progressive tilting of the depositional surface of the MV. ) l a g m ( y l a m o n a
SECTION 12 BOUGUER ANOMALY 0.
ref.level :
-110. -220. 300.
350.
400.
distance (km)
) m ( n o i t c e l f e d
FLEXURE 5000. 0. -5000. -10000. 300.
350.
400.
distance (km)
) m k ( t e e
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0. 300.
350.
400.
distance (km)
) a P G ( s s e r t s
FIBER STRESS 1. -1. 0.
300.
350.
400.
distance (km)
Figure 4.12: Continued
170
0. m
Chapter 4
1400
LITHOSPHERE EFFECTIVE ELASTIC THICKNESS (km)
s a c u L n a S e d ia n a r r e S
1300
Maracaibo Basin
Serrania de Perija
Cucuta
s d e A n i d a r e M
Barinas Basin
Bucaramanga 5
Arauca
20
1200
Medellin
r a l e i d o r C
20
0 3
Tunja
1100 Manizales
5 0
Yopal 0
1
0 1
Bogota
1000
Ibague
Villavicencio
900
l r a t e n C Guyana Shield Neiva
800
800
900
a e n d e r a a c i n a a r M r e a S L
5 0
-0
San Jose del Guaviare
1000
1100
1200
100
1300
200 km
1400
1500
Figure 4.13: Map of calculated effective elastic thickness. Note location of important strike slip faults associated to pre-Mesozoic outcrops (shaded areas: Santander-Floresta and Garzon-Quetame massifs) and location of possible mantle lithosphere subduction under the Eastern Cordillera.
9.
DISCUSSION
9.1. NEOGENE TECTONIC HISTORY AND PLATE-TECTONICS During Neogene the Mesozoic extensional basin was completely inverted (Colletta et al., 1990; Dengo and Covey; 1993; Cooper et al., 1995; Casero et al., 1995, 1997;). Upthrust of the basement blocks represented by the Garzon Massif and the Sierra de La Macaren a (Fig. 2.1) also occurred during Neogene (Van der Wiel, 1991). Inversion of extensional basins and upthrusts of basement blocks occurred by reactivation of pre-existing crustal discontinuities in response to compressional/transpressional stresses on the NW margin of the South American plate. These processes have been proposed to explain basin inversion and upthrust of basement blocks in Europe and many other areas by Ziegler et al. (1995, 1998). According to the plate tectonic history of Colombia, convergence rate of the South American plate relative to the Pacific plates was particularly fast during Late Miocene and Pliocene, triggering Andean orogeny (Cooper et al., 1995). Backarc compression resulting in inversion of rift tensional hanging-wall basins is associated with Andean-type orogens
171
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
occurs during periods of increased convergence rates between the subducting and overriding plates (Ziegler et al., 1998). This inversion of backarc rift basins results from acceleration of convergence rates between the colliding plates, their increased mechanical coupling and the transmission of compressional stresses into the backarc domain of the overriding plate (Uyeda and McCabe, 1983; Ziegler, 1993; Ziegler et al., 1998). In addition Neogene basin inversion and Andean orogeny have been correlated to the collision of the Choco Block (Cuna Terrane according to terminology of Toussaint, 1995) with the northwestern margin of South America (Duque-Caro, 1990, Cooper et al., 1995). Kerr et al. (1997) have studied the geochemistry of the volcanic rocks of the oceanic Cuna Terrane and proposed that it represents part of an oceanic plateau. Build up of intraplate compressional/transpressional stresses in the NW margin of South America during Neogene and collision of the Cuna Terrane could be favoured by impeded subduction caused by the arrival of the more buoyant oceanic plateau. Probably right-lateral transpressional deformation contributed to Andean orogeny and allowed inversion of Mesozoic normal faults. An example of a Mesozoic normal fault system, right-lateral transpressionally inverted during Neogene, is the Guaicáramo-Chámeza fault system (limiting the EC), evidenced by earthquake focal mechanisms solutions (Pennington, 1981). Some right-lateral faults ( e.g. Altamira Fault, Casero et al., 1995, 1997) were also active during Neogene.
9.2. LITHOSPHERE STRENGTH EVOLUTION AS INFERRED FROM FLEXURAL MODELLING The parameter that characterizes the apparent strength of the lithosphere is the flexural rigidity, which is commonly expressed through the effective elastic thickness (EET) of the lithosphere. The distribution of strength in the lithosphere varies vertically as a result of its thermal and compositional layered structure. However an appropriate first order description of the flexural behaviour of the lithosphere can be formulated assuming it as a thin elastic plate ( e.g.T urcotte and Schubert, 1982). The effective elastic thickness is a useful and easy parameter to estimate that provides important information on the mechanical and thermal state of the lithosphere. For an elastic plate the notion of the integrated strength is quite close to that of the flexural rigidity (Burov and Diament, 1995). These authors have shown that the effective elastic thickness of the continental lithosphere is dependent on: (1) The thermal state/age of the lithosphere (thermal age defined as a period of time required for the lithosphere to reach its present-day thermal state, assuming that the lithosphere was initially melted. The thermal age controls the depth to a specific geotherm obtained from a plate cooling model assuming that the lithosphere did not undergo thermal re-settings during this time. The thermal age gives the age of the last large-scale thermal event (Burov and Diament, 1995); (2) The coupling or decoupling state of the crust and mantle; (3) The thickness and proportions of the mechanically competent crust and mantle; and (4) The local curvature of the plate, which is directly related to the bending stress (Burov and Diament, 1995). An additional important control on lithosphere strength is the presence of pre-existing discontinuities, which are prone to reactivate reducing strength (Van Wees, 1994; Ziegler et al., 1995, 1998). The thermal state of the lithosphere controls lithospheric strength since temperature dependent creep controls the ductile strength of the lower crust and lower mantle lithosphere (Ranalli, 1995). The thickness of the mechanically competent crust and the degree of coupling or decoupling are generally controlled by composition of the upper and lower crust, total thickness of the crust, and by the crustal geotherm. If decoupling take place, as is of common occurrence in continental lithosphere, it permits as much as 50 % decrease of elastic thickness, compared with elastic thickness derived for conventional thermal profiles (Burov and Diament, 1995). The curvature of the plate depends of the rheological structure and on the distribution of external loads applied to the plate ( e.g. topography, sediment fill and plate boundary forces). Bending stresses created by major mountain belts are large enough to cause inelastic deformation (brittle failure and ductile flow) in the underlying plate, which in turn, leads to a 30 to 80% decrease of elastic thickness beneath such belts and less beneath the adjacent regions. The boundary forces and moments lead to more localized but even stronger reductions in elastic thickness (Burov and Diament, 1995).
172
Chapter 4 LEON ) m ( n o i t c e l f e d
1000.
GUAYABO 1 ) m ( n o i t c e l f e d
FLEXURE
500. 0. -500. -1000.
4000.
FLEXURE
2000. 0. -2000. -4000.
-1500. 200.
400.
600.
800.
200.
400.
distance (km)
) m k ( t e e
80.
EFFECTIVE ELASTIC THICKNESS
) m k ( t e e
60. 40. 20.
80.
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
200.
400.
600.
800.
200.
400.
distance (km)
) a P G ( s s e r t s
1. 0. -1. 400.
600.
-1. 200.
400. distance (km)
GUAYABO 3
FLEXURE
) m ( n o i t c e l f e d
2000. 0. -2000. -4000. -6000.
FLEXURE 5000. 0.
-5000.
800.
200.
distance (km)
) m k ( t e e
80.
) m k ( t e e
60. 40. 20. 0.
) a P G ( s s e r t s
400.
600.
1. 0. -1. 600.
600.
800.
800.
distance (km)
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0. 200.
) a P G ( s s e r t s
FIBER STRESS
400.
80.
800.
distance (km)
200.
400. distance (km)
EFFECTIVE ELASTIC THICKNESS
200.
800.
0.
800.
4000.
600.
600.
1.
GUAYABO 2
400.
800.
FIBER STRESS
distance (km)
200.
600.
distance (km)
FIBER STRESS
200.
) m ( n o i t c e l f e d
800.
distance (km)
0.
) a P G ( s s e r t s
600.
400.
600.
800.
600.
800.
distance (km) FIBER STRESS 1. 0. -1. 200.
400. distance (km)
Figure 4.14: 2D flexural models produced by the gradual surface-uplift o f the Eastern Cordillera during Neogene time using data from the seismic section shown in figure 4.6. L ocation of the section in Figure 4.1
173
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
GUAYABO 4 ) m ( n o i t c e l f e d
GUAYABO 5
FLEXURE
0.
-5000.
200.
FLEXURE
) 5000. m ( n 0. o i t c e l f e -5000. d
5000.
400.
600.
800.
0.
distance (km)
) m k ( t e e
80.
1000.
distance (km)
) m k ( t e e
EFFECTIVE ELASTIC THICKNESS
60. 40. 20. 0.
EFFECTIVE ELASTIC THICKNESS
80. 60. 40. 20. 0.
200.
400.
600.
800.
0.
distance (km)
) a P G ( s s e r t s
500.
) a P G ( s s e r t s
1. 0. -1. 400.
600.
1000.
distance (km)
FIBER STRESS
200.
500.
800.
FIBER STRESS 1. 0. -1. 0.
distance (km)
500.
1000.
distance (km)
Figure 4.14: Continued
9.2.1. Ef fect of l ith ospher e ther mal age on strength
One of the major controls on the effective elastic thickness of the study area is the thermal age of the lithosphere, which varies laterally. Based on this and the values obtained for elastic thickness three regions can be distinguished in the study area: Llanos Orientales (LLA) Basin. The estimated effective elastic thickness has a maximum value of 50 to 55 km in the eastern LLA basin (Fig. 4.13). The thermal age of the eastern LLA lithosphere is older; the last significant thermal event in this region probably was Cambrian rifting (Ecopetrol and Beicip, 1995, mainly in the western part of the LLA area). This correlation suggests a relatively stable strong lithosphere in this area. However in the western part of LLA basin values of effective elastic thick ness are intermediate between those of the Andean region and those of the eastern part of LLA basin ( Fig. 4.13). In the eastern part of LLA basin a strong continental lithosphere is indicated by effective elastic thickness with a constant value of 50 km for most of the Neogene and 55 km at present time (Table 4.4). The westernmost part of LLA basin is characterized by the transition between the strong (EET 50 Km) and weak (increasing EET from 10 to 25 km during Neogene time , Table 4.4) continental lithosphere. Eastern Cordillera (EC)-Magdalena Valley (MV). The effective elastic thickness in the EC has values of 25 km or less (Fig. 4.13) indicating a weak lithosphere. The age of the last large-scale thermal event related to the Mesozoic rifting phases which started 248 Ma and ended 102 Ma ago. Because the last stretching/heating event occurred during Aptian time, the estimated thermal age is about 120 ± 20 Ma. This weak lithosphere is prone to deformation as indicated by its well-developed earthquake activity (e.g.P ennington, 1981). Weak lithosphere in the area of the EC and the western part of the MV
174
Chapter 4
correlates with the area of the former Mesozoic extensional basins ( Figs. 2.13, 2.15 a nd 2.17 Chapter 2). In general weak lithosphere in the Andean region probably co rresponds to therma lly destabilized lithosphere. Young rifts and volcanic areas are characterized by low elastic thickn ess values (e.g. Ebinger et al., 1989, in Burov and Diament, 1995) due to thermal weakeni ng and necking of the lithosphere. Stretched lithosphere is weak and low elastic thickness values in the EC may have inherited this weakness. Similar results were obtained by Van der Beek and Cloetingh (1992), who demonstrated that preceding thermo-tectonic events explain the present-day flexural response of the lithosphere in the Betic cordilleras of Spain. Central Cordillera (including westernmost MV). Weakest lithosphere in the area of the Central Cordillera (EET values of 5 km, Fig. 4.13) also correlates to the recent volcanic arc in the Central Cordillera. The thermal age of the Central Cordillera is very small since at the present-day there is active volcanism on this mountain range. Volcanism and very low elastic thickness suggest a hot and weak lithosphere in the area of the Central Cordillera. 9.2.2. Coupli ng/decoupli ng state of t he crust-mantl e, thi ckness and pr oporti ons of mechani cally competent cru st an d man tle
Burov and Diament (1995) have compared analytical and numerical solutions for the flexural behaviour of the lithosphere for both the elastic model and a stratified yield stress envelope model with realistic brittle-elasto-ductile rheology. In order to compare results of both models deflection of the brittle-elasto-ductile plate was also modelled by these authors as the deflection of an equivalent elastic plate with space and temporally variable effective elastic thickness. Quantitative relationships between thermal age of the lithosphere, effective elastic thickness and crustal thickness showing the reduced elastic thickness effect produced by crust-mantle decoupling obtained by Burov and Diament (1995), are shown in Figure 4.15. These authors showed that the lithosphere is decoupled when the crustal thickness is greater than a critical value, and coupled, when the crust is thinner than such critical value. The critical value corresponds to the transition zone in Figure 4.15. It increases with age of the lithosphere until it reaches an asymptotic value close to 35 km for ages greater than 750 Ma. Plotting the thermal age, crustal thickness and obtained elastic thickness for the EC, Central Cordillera and LLA area for different times during the Tertiary and present-day indicates a relatively young lithosphere and crust-mantle decoupling ( Fig. 4.15). Thus Tertiary pre-Andean and Andean deformation of the upper crust of the EC was largely independent of upper mantle deformation. These zones were separated by ductile flow in the lower crust. In areas of young lithosphere like the Alps ( e.g. Okaya et al., 1996) the elastic thickness estimates (5-30 km) are smaller that the depth of the base of the mechanically strong upper crust, corresponding to the isotherm 200° -300° C (Burov and Diament, 1995). The similar age and elastic thickness obtained for the Colombian EC suggests a similar picture for the EC. According to Burov and Diament (1995) for very young lithosphere, as is the case for the Central Cordillera and the EC during Palaeogene time, the thickness of the competent crust may be greater than, or comparable with, the thickness of the competent mantle lithosphere. Consequently, the integrated strength of the young plate is controlled at a large extent by the strength of the crust (Kusznir and Karner, 1985; Burov and Diament, 1995). The mantle lithosphere perhaps does not contribute to the plate strength in these cases. For ages larger than 100-150 Ma, the lithospheric strength is significantly controlled by the strength of the mantle (Burov and Diament, 1995). If the thermal age of the EC lithosphere is about 120 Ma, probably mantle began to contribute to lithospheric strength only since Neogene time, as suggestedby the obtained elastic thickness values. According to Ziegler et al. (1998) compressional deformation restricted to crustal levels involves “simple shear” type detachment of the crust at the level of the rheologically weak lower crust from the mantle lithosphere. This mechanism probably occurred in the EC area during Palaeogene and beginning of Neogene time, as suggested by reduced EET values at those times.
175
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
Thermal Age of Lithosphere (Ma) 0
25 0
50 0
10
7 0
8 0
o 4
5 8, 9
h
9
3
62
e
7
7 1 Central
30
o
Cordillera 3 0
M
E E T ( k m )
2 6
1 6 2
4 0
2 0
1 0
40
1 1 0
T r a n s i t i o n z o n e
8
5
3
1 0 0
4
20
k
9 0
C O U P L I N G
5 0
6 0
7 0
D E 10 C 10 O 10 U Llanos P L Basin I N G
7 3 8, 9 5 4
End Cret aceous End Paleogene
End Neogene 3
Inv erted basi n compartm ent
Figure. 4.15: Coupling/decoupling state of the lithosphere as a function of the age of the lithosphere and the Moho depth. One can see that for each age there is a critical value of crustal thickness (transition zone curve). If the crustal thickness is larger than this critical value, the lithosphere is practically always decoupled, resulting in low EET. In turn, thin crust results in mechanical coupling between the crust and mantle and in high EET values. For a lithosphere older than 750 Ma the value of the critical crustal thickness is practically constant and equal to 35 – 40 km (from Burov and Diament, 1995). Colombian data: numbers atached to a symbol (circles, squares and crosses) indicate basin compartments shown in Fig 2.18. Numbers 2 to 9 refer to basin compartments the former Mesozoic extensional basin in the area of the EC
176
Chapter 4 9.2.3. Ef fect of local cur vature of th e plate on l ith osphere str ength
Radius of plate curvature, which is related to the gradient of the bending stress within the plate, is a parameter that also controls effective elastic thickness according to Burov and Diament (1995). These authors calculated the relationship between the radius of curvature, the thermal age and the equivalent elastic thickness. Figure 4.16 shown their result for a case of crust-mantle decoupling. Lithospheric flexure may be responsible for localized reductions of the effective elastic thickness that may be between 0.1 and 0.5 orders (Burov and Diament, 1995). In their figure there are two families of curves: grey curves represent the concave downward flexure, with compression in the uppermost crust and mantle, while the black curves represent concave upward flexure, with extension in the uppermost crust and mantle. The downward flexed plate generally cannot be so highly flexed as the subducting plate, because the stresses required for the concave downward flexure are much higher than those for the concave upward flexure. Plotting the respective values obtained for the EC ( Table 4.5 and Fig. 4.16) may explain the drastic reduction in strength (modelled assuming the end of a broken plate (elastic thickness = 0 under the Cordillera) in the place where the continuous plate model under the Cordillera predicts maximum curvature and maximum bending stress. Also the reduction of elastic thickness in the Middle Magdalena Valley may be at least partially explained by this effect. Curvature
Radius curvature
Antiform min Antiform max Synform min Synform max EET: effective elastic thickness.
EET (km) 15 50 15 50
Bending stress (GPa) 0.60 0.20 1.20 0.65
EET/ Bending stress (km/GPa) 25.00 250.00 12.50 76.92
Radius (km) 0.625 6.250 0.313 1.923
Log10(radius) -0.20412 0.79588 -0.50515 0.28400
Table 4.5: Estimated radius of curvature of the flexed lithosphere under the Eastern Cordillera calculated from the modelled bending stress.
EC
Log10 (Radius of curvature (Km)) Figure 4.16: Decoupled lithosphere (lower crust with low temperature of creep activation and/or young plates). The decoupled model fits most of the continental data excepts few cases where the lithosphere is likely to be coupled (Himalaya). The lithospheric flexure can dramatically decrease the effective elastic thickness of the lithosphere. (from Burov and Diament, 1995). Colombian data: LL Llanos basin. EC: Eastern Cordillera.
177
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes 9.2.4. Ef fect of pr e-exi stin g disconti nu iti es on l ith osphere str ength
Van Wees (1994) has shown the important role of pre-existing discontinuities is reducing lithosphere strength. I have discussed this in Chapter 3. This effect is probably very important in the case of the EC since the leading mechanism of orogeny has been basin inversion with compressional reactivation of former extensional faults. The reactivation of faults also suggests an oblique orientation of the compressional stress trajectories, which is in agreement with oblique plate convergence. Stretched lithosphere is weak in part due to those discontinuities and low elastic thickness values in the EC may inherit this weakness. Additionally in the EC values of estimated effective elastic thickness less than 10 km occur in the northern and southern parts of the EC, where major strike-slip faults such as the Santa MartaBucaramanga, Bocono and Altamira faults are located. Regionally Andean deformation affected weak lithosphere. However, the weakest lithosphere in the EC seems to be associated with strike-slip faults (Fig. 4.13). Such faults probably ar e deep enough into the lithosphere to reduce its total strength. Outcrops of pre-Mesozoic igneous and metamorphic basement rocks also are associated with strike-slip faults and weak lithosphere ( Fig. 4.13). Probably such faults also have vertical components that brought basement rocks to surface. 9.2.5. Other effects on l it hospher e str ength
Elevated topography of mountain ranges creates normal stresses high enough to cause significant flexural deformations and bending stresses on the underlying lithosp here (e.g. Burov and Diament, 1995). Inelastic deformation below the EC is suggested by the elevated irrealistic bending stresses greater than rock strength estimated from flexural models (Figs. 4.10 and 4.11). Burov and Diament (1992, in Burov and Diament, 1995) have demonstrated s trength reduction under mountain ranges. The better fit obtained f or a broken plate end elastic model under the EC (Fig. 4.12) than for a continuous plate (Fig. 4.11) suggests a dramatic strength reduction (to zero) under the Cordillera and a possible “subduction” under the EC as hypothesized by Colleta et al.( 1990). The presence of rhyolitic and andesitic volcanic rocks of Neogene age near Paipa and Iza (Renzoni et al., 1967; Romero and Rincon, 1990) demonstrates Neogene magmatic activity in the EC at least locally. Additional thermal heating from below (evidenced by the Paipa-Iza volcanic rocks) also contributes to a reduction of lithospheric strength under the EC. In zones of collision, the lithosphere is subjected to significant plate boundary loads (Burov and Diament, 1995, Ziegler et al., 1998) creating significant local strength variations (Burov and Diament, 1995). Although such plate boundary forces may have existed in the flexural history of the EC horizontal stresses may also have a significant role on the flexural history of the EC and on the lithosphere strength (Cloetingh, 1988; Burov and Diament, 1995). However, in the present situation models do not require plate boundary forces or subsurface loads, and for simplicity I neglected their effect. 9.2.6. Tempor al chan ges of l it hosphere str ength
In the weak EC region lithospheric strength increased during Neogene time (Table 4.4). Cooling and lithosphere thickening after the Mesozoic rifting probably dominated thi s increase in strength, although other mechanisms also participated as already discussed. Pre-existing discontinuities also decrease lithosphere strength. The effective elastic thickness obtained in the EC region gradually increased from 10 k m at the end of Middle Miocene during deposition of the top of Leon Fm up to 25 km at present-time (Table 4.4). During Palaeogene time small EET values (<10 km) imply that lithospheric strength was only represented by the upper crust ( c.f.C loetingh and Burov, 1996). Gradual increase of EET values during Neogene suggests a gradual participation of the mantle lithosphere in the total lithospheric strength ( c.f.C loetingh and Burov, 1996).
178
Chapter 4
The progressive increase in EET values in the Andean region and constant EET in the LLA region imply a general increase of lithospheric strength. Apart of the apparent bulge migration because of convergence of the foreland toward the mountain belt, such general increase of lithospheric strength would produce an increase in width of the foreland basin and migration of the flexural bulge away f rom the load. In gener al this trend is observed in the LLA basin during the whole Tertiary ( Figs. 3.4, 3.10, 3.12. 4.2 , and 4.4 and Table 4.4). In the LLA the Paleocene basin was narrow (<100 km) and th ere is evidence in its eastern part for erosio n of the Cretaceous sedimentary record probably due to erosi on of the u plifted flexur al bulge. Later during Palaeogene the LLA foreland width increased to approxim ately 300 km at the en d of deposition of Carbonera Fm (end of Early Miocene), and up t o a maximum of appr oximately 400 km in th e north during deposition of Guayabo Fm in Neogene tim e (Figs. 3.4, 3.10, 3.12, 4.2, 4.4 and 4.8). Dur ing the whole Tertiary maximum width of the LLA basin is located in the northern Arauca area and decreases toward the south suggestinga so uthward decrease in lithospheric strength as also indicated by flexural mode ls (Figs. 3.4, 3.10, 3.12, 4.2, 4.4 and Table 4.2). In the southwestern LLA basin close to the Serrania de La Macarena ther e is seismic evidence of partial erosion of the Neogene Guayabo F m (Fig. 4.7). Seismic lines of other part s of the basin (e.g. Fig. 4.6) do not show evidence of similar er osion, which affects only this part of the basin close to the Serrania de La Macarena. Probably this was the result of flexural rebound ind uced by an er osional decrease in topographic load in that part of the EC (Garzon-Quetame Massif, Fig. 2.1). The Serranía de La Macarena can be regarded as a foreland upthrust block. Because in this SW part of the basin this erosion of Guayabo Fm implies a reduction in sediment thickness and estimated subsidence the flexural deflection, the width of the basin and EET value have been underestimated for the southernmost modelled profile. In the westernmost part of the LLA basin the calculated EET value is lower than the 50 km estimated for the eastern continental lithosphere. In this region model results ( Table 4) indicate that the zone of weak lithosphere (characterized by a reduced elastic thickness 50-km) moved craton-ward possibly as the result of heat transmission from the “hot” Andean region to the “cold” cratonic region. The characteristic time of these thermal relaxation processes is about 50 Ma (McKenzie, 1978). An alternative explanation may be found in the flexural behaviour of a viscoelastic plate. In such model stress relaxation leads that a load initially isostatically compensated by a regional flexural deflection with time to evolve to local isostatic compensation. If a thin elastic plate model is applied in this case for several time intervals, the EET will decrease with time (Walcott, 1970; Watts et al., 1982; Garcia Castellanos, 1998). Characteristic time of this viscous stress relaxation is 1 Ma. Taking into account the Neogene time interval studied (16.4 Ma) the viscous stress relaxation hypothesis is the most probable explanation. Additionally most of the thermal relaxation occurred during Palaeogene time (48.6 Ma). 9.2.7. Suggesti ons for fu r ther studi es
Detailed stratigraphic studies of the Neogene sediments, their relations with deformation, and fission track data are needed to accurately constraint models about the deformation and uplift history of the EC.
10. CONCLUSIONS Comparison of the observed Neogene tectonic subsidence with the remaining thermal subsidence assuming tectonic quiescence indicates that the Mesozoic extensional basin was completely inverted during Neogene time to form the EC. Increased South American plate convergence rate and possibly collision of the oceanic plateau Cuna Terrane favoured this compressional/transpressional deformation. Flexural subsidence produced by the topographic load of the EC explains the observed Neogene subsidence in the LLA and MV. Irrealistic high bending stresses predicted by flexural models under the EC and the better fit obtained by flexural models assuming a broken plate under the EC, support the hypothesis of mantle lithosphere subduction (Colletta, et al., 1990) under the EC. Flexural model results for different stratigraphic markers within the Neogene sedimentary record, and comparison with available fission track data and other evidence of deformation/rock-uplift, suggest that during Miocene time important surface-uplift (> 1000 m) occurred mainly at the margins
179
Neogene Basin Inversion History of the Eastern Cordillera, Colombian Andes
of the EC. In the middle axial Tunja-Sabana de Bogotá region, where the Palaeogene sedimentary record has been partially preserved, important surface-uplift occurred later during Pliocene time as recorded by the exceptional palynological record of the Sabana de Bogotá (Helmens, 1990; Andriessen et al., 1993). Such uplifting history can be explained in terms of basin inversion: contractional reactivation of Mesozoic extensional faults initiated during Palaeogene times leading to an initial slight basin inversion; then during Neogene time the compressional deformation increased in rate and magnitude, leading to a complete inversion of the original extensional basin. Scarce fission track data indicate exhumation and the results of flexural models suggest that moderate (< 1000 m but mostly ≤ 200 m) and local surface-uplift of blocks, mostly associated with the master normal fault system that delimitated the Mesozoic extensional basins, initiated during Palaeogene times (see Chapter 3). However, during Neogene times increased surface-uplift (> 1000 m) affected the former Lower Cretaceous depocenters adjacent to the master normal fault systems delimiting the basin. Complete inversion of these master normal fault systems led to “extrusion” of the sedimentary fill of half graben basins that now form the eastern and western flanks of the EC. Regional structural balanced crosssection interpretations ( e.g. Colletta et al., 1990; Cooper et al., 1995) suggest that maximum dip-slip displacement and shortening occurred on the thrust faults connected to inverted Mesozoic normal faults that now approximately delimit the Lower Cretaceous outcrops of both flanks of the EC. In these uplifted flanks Lower Cretaceous or older exposed rocks indicate that all Late Cretaceous or younger sediments have been eroded. However, only in the Santander, Quetame and Garzon massifs all the sedimentary cover has been removed by erosion. Proximity of these massifs to important strike-slip faults suggests that inversion was associated with important strike-slip motions. Preservation of the Palaeogene sedimentary record in the axial Bogotá-Tunja zone of the EC, as well as the Neogene sedimentary record of the Sabana de Bogotá area suggest that this axial region remained low during Miocene time and was uplifted until the Pliocene. One of the major controls on the effective elastic thickness of the study area is the thermal age of the lithosphere. Based on the thermal age and the values obtained for elastic thickness, three regions can be distinguished in the study area: (1) The LLA basin where effective elastic thickness has a maximum value of 50 to 55 km and the thermal age of the eastern LLA lithosphere is Palaeozoic (this correlation suggests a relatively stabilized strong lithosphere in this area); (2) The Eastern CordilleraMV, where effective elastic thickness has values of 25 km or less, indicating a weak lithosphere and the thermal age is about 120 +-20 Ma (this weak thermal destabilized lithosphere is prone to deformation; weakness there is inherited from the former Mesozoic extensional basins); (3) The Central Cordillera (including westernmost MV) with EET values of 5 km and very young thermal age as indicated by a recent volcanic arc in the Central Cordillera. EET values lower than 10 km occur in the northern and southern parts of the Eastern Cordillera, where major strike-slip faults such as the Santa Marta-Bucaramanga, Bocono and Altamira faults are located. Regionally Andean deformation affected weak lithosphere; however, the weakest lithosphere is associated with strike-slip faults. Such faults probably reach deep enough into the lithosphere to reduce its total strength. Results of flexural modelling indicate crust-mantle decoupling. In the Eastern Cordillera mantle began to contribute to lithospheric strength probably only since Neogene time. Existence of pre-existing discontinuities probably significantly reduced lithosphere strength. Maximum radius of curvature of the flexed plate under the Eastern Cordillera also reduced its strength. In the weak Eastern Cordillera region, lithospheric strength increased during Neogene time. A local lithospheric weakening effect in the western LLA basin can be interpreted as viscous stress relaxation.
180
CHAPTER 5 MAP VIEW RESTAURATION OF TRANSPRESSIONAL BASIN INVERSION IN THE EASTERN CORDILLERA 1. INTRODUCTION In areas of compressional tectonics, the traditional method for restoring the deformation is to use balanced cross-sections (Dahlstrom, 1969). Such restoration takes into account both folding and faulting. The main assumption is conservation of lengths or surface areas of the layers in the crosssection. This assumption is not valid if there are differential motions of material in directions perpendicular to the plane of the section, such as occur in strike-slip faulting or block rotations about vertical axes. This restriction makes the methods of d oubtful use in areas of non-planar deformation such as the Northern Andes and, in particular, the EC ( Fig. 1.1), which is located in a broad z one of defor mation resulting from the interaction of the Sou th American, Nazca and Caribbean plates ( Fig. 5.1). Such interaction has fragmented the NW corner of the South American plate into a number of micro-plates or tectonic blocks ( e.g. Kellogg et al., 1985; Freymueller et al., 1993). Relative movements of the micro-plates has resulted in deformed belts with a combination of compressional thick-skinned ( e.g. Julivert, 1970; Colletta et al., 1990), thin-skinned thrusting (Dengo and Covey, 1993; Fajardo-Peña, 1998), folding (Campbell and Burgl, 1965; Julivert, 1970; Kammer, 1999a, b) and strike-slip faulting (Campbell, 1968; Kammer, 1993a, 1996, 1999 a,b; Acosta et al., 2000) which form several mountain ranges or strike-slip fault zones in the Northern Andes. The variable orientation of these deformed belts, as well as the combination of compression and strike-slip movements, implies non-planar deformation, which cannot be, restored properly using balanced cross-sections alone. Some algorithms and commercial computer programs have been developed for 3-D structural restoration. Rouby (1995), Rouby et al . (1998) among others have developed algorithms for unfolding non-cylindrical folds. They allow unfolding individual blocks. Numerical procedures for fitting blocks in plan view, applicable to non-planar deformation, have been developed for regions of continuous deformation (Cobbold and Percevault, 1983) and adapted to discontinuous deformation of regions dominated by strike-slip motion (Audibert, 1991), normal faulting (Rouby et al ., 1996) and reverse faulting (Bourgeois et al ., 1997). However, all these methods require in addition to balanced structural sections, structure contour maps of the layer to be restored. In the case of the EC such methods have up to now been impossible because structure contour maps do not exist for the EC. These methods would require a detailed knowledge of the 3-D structure, which is beyond present-day knowledge of this mountain range. In this Chapter I use a modified method for map view structural restoration that does not require structure contour maps and applied it to restore the Neogene Andean deformation of the EC. The aim of this chapter is to restore in map view the Neogene Andean deformation of the EC and neighbouring areas (Figs. 1.1 and 5.1) using all data available in literature. I compiled more than 50 balanced cross-sections and used the am ounts of shortening estimated in those cross-sections. Understanding the Neogene compressional Andean deformation mechanisms is important to understand development of petroleum structural traps during the Neogene and the preservation of early formed traps.
2. TECTONIC SETTING 2.1. PRESENT DAY PLATE-TECTONIC SETTING The Colombian Andes lie in a complex tectonic area of convergence between the South American, Nazca and Caribbean plates ( Fig. 5.1, e. g. Freymueller et al ., 1993). Kellogg et al . (1985) have interpreted the tectonics of the region in terms of two a dditional micr oplates, Panama and the North Andes, both of which are subject to internal deformation ( Fig. 5.1). At present, subduction (70 mm/yr) of the oceanic Nazca Plate occurs at the western margin of South America, and the trench is seismically active ( Fig. 5.2; Freymueller et al ., 1993). The North Andes are bounded by the Colombia-
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Colombian Basin South Caribbean deformed belt
CARIBBEAN PLATE
lt lt e B Be d d a a r m rm f o fo e d de u i n S t o
North Panama fold belt
PANAMA BLOCK
Maracaibo
e s n d A i d a e r M
in c a J
n a S
CHOCO BLOCK
NAZCA PLATE
Block
Venezuela A R E L L I D R C O N R E T S Colombia E A
SOUTH AMERICAN PLATE
A I B M O L O C
h n c e r t o r d a c u E i a b o m l o C
N T E R E S W
Detail in Fig. 5.3
PLATES
F O
S A N E R R T E
Continental plate Oceanic plate
DEFORMATION ZONES
I C A N E C O - O
Eastern Andes
E L A A P
Accreted oceanic terranes
0
500 km
ESTRUCTURES Thrust fault Strike-slip fault Normal fault
Figure 5.1. Present day plate-tectonic setting and major structures of the Northern Andes in the NW corner of South America.
Ecuador Trench and Panamá Block on the west, the South Carib bean deformed belt to the north, and the Boconó and East Andean fault zones to the east ( Figs. 5.1 and 5.2). Within the northern part of the Andes, significant displacements have occurred on the Oca (right- lateral, east-west trending) and Santa Marta-Bucaramanga (left-lateral, northwest-southeast trending) fa ults (Fig. 5.2; e.g. Campbell, 1968; Ir ving, 1971; Schu bert, 1982). The Boconó Fault, Santa Marta-Bucaramanga Fault, and Oca-El Pilar faults (Fig. 5.2) define the wedge-shaped “Maracaibo Block’ which has moved northward relative to sta ble South America during late Cenozoic (Bowin, 1976; Mann and Burke, 1984; Pindell and Erikson, 1993). On the east side of the North Andes, right-lateral transpressive movement is taking place on the Boconó and east Andean (Guaicáramo) fault systems ( Fig. 5.2, e.g. Pennington, 1981). Th e Boconó Fault in the Mérida Andes of Venezuela is mainly a right-lateral strike-slip fault ( Fig. 5.2, Schubert,
182
Chapter 5 CARIBBEAN PLATE
Colombian Basin South Caribbean deformed belt
North Panama fold belt
Panama Block
k c o l B o c o h C
NAZCA PLATE CALI
r o a d u c E i a b m l o o C
h c n e r T
Oca fault
t l e B B t d l e u S e B r m c a a r a n u d f o n e i m ta U S m d e a M r r a n o t o g a r m f e i n a ta d c i t F a l m a a J u ra t e fa lt e s u y lt n o m s lt a S R u f a BUCM
El Pilar Fault
Maracaibo Block
s d e A n a i r d l t M e f a u
n o c o o B
0
500 km
l t a o n r F ) o m r a EC i c a t e m u BOGO G y s ( S n l t VILL a e a u d F n t A a s E
PAST
SOUTH AMERICAN PLATE ESME JERS
LA T A
PLATE VELOCITY VECTORS WITH RESPECT TO STABLE SOUTH AMERICA (from Kellogg et al., 1985 and Freimueller et al., 1993) RELATIVE DISPACEMEN T OF SEVERAL GPS STATIONS WITH RESPEC T TO VILLAVICENCIO STATION (S TABLE S OUTH AMERICA)1991 TO 1994 (from Mora, 1995, in Acosta et al., 2000)
t l a u F z a i t i n l u l L a
F g a n t a l a P a
20 mm/yr Plate velocity vector relative to South American Plate Plate velocity vector relative to the North Andes Block
ESTRUCTURES Thrust fault Strike-slip fault Normal fault
Figure 5.2. Tectonic features and boundaries of the North Andean Block. All plate velocity vectors are relative to the South American plate, except the grey vector, which represents the Caribbean-North Andes vector of Kellogg et al . (1985, modified by Freymueller et al ., 1993). Relative displacement measurements (mm/year) from 1991 to 1994 of several Global Positioning System (GPS) stations located within the Andes with respect to the Villavicencio station located in the Llanos Basin (stable South American Plate). (modified from Mora, 1995 in Acosta et al., 2000).
1982). Sub-parallel thrust faults flank the Mérida Andes suggesting significant compression normal to the Boconó Fault (Henneberg, 1983). The East Andean fault system (faults limiting the eastern border of the EC) consists of west-dipping reverse faults ( Fig. 5.2). Based on a study of focal mechanisms Pennington (1981) showed that transpressive right-lateral slip is occurring on these faults, and that the North Andes Block is moving NNE relative to the South American Plate. To the south the Ecuadorian Andes include a system of sub-parallel northeast-trending right-lateral strike-slip faults and northtrending thrust faults ( Fig. 5.2). The relationships between these faults are not well understood. Focal mechanism studies suggest thrust faulting on a N-S trending fault as well oblique transpressional
183
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
displacement (Lyon-Caen et al ., 1990). Based on observed slip rates, Kellogg et al . (1985) predicted that the North Andes relative to South American Plate is moving approximately 10 mm/yr toward the NE (Fig. 5.2). The Southern Caribbean Plate boundary is generally accepted to be a distributed zone of deformation, although there remains disagreement as to how much of the North Andes should be considered “Caribbean”. The principal part of the deformation zone consists of a series of deformed belts along the Caribbean coast of South and Central America (Pindell and Barret, 1990; Mann et al ., 1990). In Colombia the Southern Caribbean deformed belt includes the Sinú and San Jacinto belt s (Fig. 5.2). Kellogg et al . (1985) used a block vector model to esti mate the convergence rate between the Caribbean and North Andes to be 17 mm/yr to the southeast ( Fig. 5.2). Dewey and Pindell (1985) found a similar convergence between the Caribbean Colombian Basin and the Maracaibo Block ( Fig. 5.2).
3. STRUCTURE OF THE EASTERN CORDILLERA In the EC ( Figs. 1.1, 5.1 and 5.3) geological mapping is largely based on photogeologic interpretations. Str uctural field studies are scarce and seismic data is available only for the axial region and eastern and western foothills. Frequently seismic data are of poor quality (Dengo and Covey, 1993). Structural interpretations are thus based on limited data and conceptual models. All structural interpretations recognize reverse faults limiting the mountain range in its eastern and western borders. Earlier interpretations (Campbell and Burl, 1965; Julivert, 1970) emphasized high-angle faults and assume a dip increase with depth for the fault surfaces. In contrast most recent structural interpretations based on local seismic data (Colletta et al ., 1990; Dengo and Covey, 1993; Cooper et al ., 1995; Roeder and Chamberlain, 1995) emphasize thin-skin deformation and flattening at depth. Presence of both types of faults has been assumed in other interpretations (Kammer, 1993b, 1996, 1999a, b; ESRI and Ecopetrol, 1994; Linares, 1996; Kammer and Mojica, 1996; Fajardo-Peña, 1998). Based on outcrop observations Campbell and Burgl (1965) and Julivert (1970) recognized the general double vergence of the EC. They recognized two main thrust systems developed in the margins of the Cordillera (Eastern Guaicáramo thrust system limiting with the LLA basin and western La Salina thrust system limiting the MV Basin; Fig. 5.3) and i dentified the main thrusts. They recognized some asymmetry in the EC, for example the Quetam e Massif (Fig. 2.1) is only present in the eastern margin of the mountain range. In the western margin they recognized low angle east-dipping thrusts parallel to the mountain front, with a series of splays fr om the mountain front extending into the Magdalena Valley at a 20 º angle ( Fig. 5.3). The splays pass into north-plunging anticlines (Campbell and Burgl, 1965). These faults are in an en-échelon arrangement (Julivert, 1970). Within the Middle Magdalena Valley the structures are gentle and consist of a series of anticlines (containing oil), usually associated with reverse faults. These structures tend to converge in a southerly direction with the La Salina Fault (Fig. 5.3; Julivert, 1970). Julivert (1970) suggested that the Eastern Cordillera shows a fan structure, with some similarity to that of the Pyrenees and differentiated the mechanically different behaviour of the heterogeneous Precambrian and Palaeozoic metamorphic and sedimentary mechanical basement from the Mesozoic and Cenozoic sedimentary cover. In fact he suggested that the main structural features of the chain are determined by the relationships between basement and sedimentary cover: (1) Passive behaviour of undeformed tilted cover on basement blocks or simple draping structures occur in regions where the thickness of the preserved sedimentary cover is small; (2) Complex tight structures, where the basement has less rigid behaviour occur in areas where the thickness of the preserved sedimentary cover is larger; (3) Independent behaviour of the cover, due to gravity (small-scale collapse structures driven by gravity and favoured by erosion), t o salt (salt in anticlines but without clear evidence of typical salt domes in the Sabana de Bogotá area ; Fig. 2.1 f or location), and to disharmonic relationships occur in areas where the thickness of the preserved sedimentary cover is large ( e.g. axial region Sabana de Bogotá). In these regions the décollement of the cover from its basement can be facilitated by thick shales and accompanied by horizontal translation. The basement crops out within the mountain range in two fault-bounded massifs arranged en-echelon: the southward-plunging Santander-Floresta and the northward-plunging Garzón-Quetame massif s (Fig. 5.3). A rather narrow strip between the two massifs in the
184
Chapter 5 1700
Sierra Nevada de Santa Marta
1600 Lower Magdalena Valley
1500
i d a r e M
s d e A n
1400
TA
1300 PP SO M
N h F C
r a l e l i d r o C
1200
B F
OP
S a F F C
A C
h D F
1100
o F S
F T C F
1000
F G
B F
LLANOS ORIENTALES BASIN (LLA) 90 0
0 l a tr n e C
100 km
1.200.000
1.300.000
Neogene structures
80 0
F A
Paleogene or older structures
e a d n e a r i a n c a a r r M e a S L
70 0 Guyana Shield
800
90 0
1000
1100
1200
ESTRUCTURE S Thrust Strike-slip fault Normal fault fault Anticlin Synclin e e 1300 1400
Figure 5.3. Structure map of the EC and surrounding areas. (from Geotec, 1976; ESRI and Ecopetrol, 1994) and subsurface maps of the Llanos (Ecopetrol and Beicip, 1995), Magdalena Valley (Ecopetrol et al., 1994; Geotec, 1994) and Maracaibo basins (Roure et al., 1997). In the Llanos, Magdalena Valley and Maracaibo basins Palaeogene structures have been recognized in the subsurface below Neogene sediments. BF Bucaramanga Fault. Fsa La Salina Fault, FB Bituima Fault, FT Cusiana-Tamara Fault, FCh Chucurí Fault, FC Cambras Fault, FDh Dos Hermanos Fault, FG Guaicáramo fault, FSo Soapaga Fault, FA Altamira Fault, NM Nuevo Mundo Syncline, AC Arcabuco Anticline, PP Payóa-Provincia Anticline, OP Opón Oil Field, SO La Salina Oil Field, CF Cupiagua Oil Field.
185
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
axial part of the range, shows a more tightly folded synclinorium structure where Tertiary rocks have been preserved (Julivert, 1970). Based on surface geology and limited seismic data Colletta et al . (1990) concluded that the Cordillera could be interpreted as the result of inversion of Jurassic-Cretaceous basins during the Miocene-Pliocene shortening event (Andean orogeny). Reactiv ation of normal faults could have induced ramp anticline structures such as the Arcabuco Anticline ( Fig. 5.3). They recognized that the opposite vergence of the mountain ra nge is not symmetric. The amount of shortening they estimated from a regional balanced cross-section ( Fig. 5.4) is 105 km (L1/L0 66%). In a later paper (Colletta et al ., 1995) suggested based on analog ue models that in the Llanos foothills ( Fig. 5.3) backthrusts and forward thrusts developed at the same time and that their intersection corresponds with a potential décollement horizon. They suggested that due to the presence of a regional décollement level at the base of the Carbonera Fm and to intense erosion, triangle zones and passive roof duplexes could develop along the Andean foothills. Tectonic inversion could have produced “short cut” low-angle faults while some of the existing normal faults are expected to have been passively transported without reactivation. Dengo and Covey (1993) extrapolated two dominant styles of deformation to the entire mountain range: (1) Basement-detached fold-and-thrusts, characterized by low-angle thrust faults and fault-ramp anticlines that involve Jurassic through Tertiary rocks. They assume, without proof, that the geometric models of fault-bend and fault-propagation folding (Suppe, 1983; Suppe and Medwedeff, 1990) are applicable to the whole chain; (2) Basement-involved reverse faults, many of which have a high-angle dip at surface. They propose that these basement-involved reverse faults were active during Pliocene-Pleistocene and displace earlier (Oligocene? to Miocene) thrust faults. Their regional palinspastic restoration suggests approximately 40% shortening (150 km) in the sedimentary cover and one third of this value as basement shortening. In order to explain the difference in shortening between basement and cover they proposed as responsible for basement shortening a hypothetical low-angle mid-crustal detachment that extends westward to the Central Cordillera up to the western edge of the South American Plate. They speculated without evidence the existence of a decapitated graben below the basement rocks of the MV. Based on the assumption that the whole Cretaceous sedimentary record represents a thermal sag basin fill following Jurassic rifting and ignoring earlier work by Fabre (1987, who demonstrated the existence of an Early Cretaceous extensional basin), Roeder and Chamberlain (1995) reconstructed the geometry of the Cretaceous basin assuming it originated from the flexural subsidence produced by a line load. They assumed that this flexural model was wholly responsible for the thickness variations within the Cretaceous, and positioned stratigraphic columns reported by Colletta et al . (1990) within this template. As a result of these assumptions they constructed a more than 350 km (their figure 6) long cross-section for the Cretaceous basin that, compared to the present-day width of the EC, would imply 230 km of shortening. They extended downward the Dengo and Covey (1993) structural section assuming low-angle thrust faults affecting the whole upper crust fitting a Moho geometry produced by a line loading flexural deflection. They postulated that a polyphase history involving foreland upthrusts redeforming the fold-and-thrust belt is unlikely and suggested that, if anything, foreland upthrusting occurred prior to the main deformation event. Their speculative interpretation shows three major basement-involved thrust sheets in the Cordillera. The shallow, westward-dipping slices imbricate basement and sedimentary cover and truncate pre-existing rift geometry. Based on surf ace geology and limite d seismic data Cooper et al . (1995) returned to a rift inversion interpretation (Fig. 3.37 Chapter 3). Their shortening estimate is 68 km. They also recognized thickening of Jurassic, Early Cretaceous and proba bly pre-Jurassic (?) sediments in the hanging-walls of the Guaicáramo, Arcabuco (Soapaga Boyacá) a nd La Salina fault systems ( Fig. 5.3) and concluded that the EC structure is the result of the inversion of a Jurassic-Cretaceous basin. They also suggested that the Cusiana-Támara eastern frontal thrust faul t (Fig. 5.3) system originated as a Late Cretaceous normal fault and continued to be episodically active as a normal fault until mid-Miocene, in order to accommodate flexural loading produced by compression and surface-uplift of the Central Cordillera and the EC. They also recognized shallow thin-skinned thrusting.
186
Chapter 5
e h t n i k c o l b c i d t o e l e e s z d b n u o n d s t u e e e o e n e o n g e M m b m e o c e t t y e l r l a a o g o l r s u f e a a e e a r a B F d N P C E
m k 0 0 1
0
E S n i s a B s o n a l L F l a p o Y F o m a r a c i u G
n o i t a m r o f e F a c s e P d r e t f a h t m g k n e 9 l n 1 . t n A o c u b a c r A o 2 i t c = e s 0 k c L o l b d e . t n A z e l e V d n u o b t l u a F a n i l a S a L F
n o i t a m r o f e d e r o f e b h t g n e l n o i t c e s k c o l b d e d n u o b t l u a F
f L
m k 5 0 1 + m k 9 1 2 = 1 L
a n e y l e a l d l a g a V M 0
4 1 m K
W N
0 8 m K
187
a i n f o p n n e o e i s w t r e e b v n i h t y g n b e l d e y t a a r d e t n n e g e s e e r g P n : a r 1 L n . i a s t e n n i u l o n m i p e h n t e e f o t w y e r t e b h t m g o n e e g l e n c o i n t e a g r m e r v f o e e l d b u e r o d P : e 0 h t L . g n ) 0 i t 9 a 9 r t 1 s , . u l l l i a e C t E t . a 8 t e . e 5 h t l l e o f o C r u g n i m o i F o t r n c f ( i e s . n n i s o i s s t o a c r e c b s l d e a e h n t c o n f s a i o l n a t e n o B x i t . e a 4 . i c c o 5 o L z e . r o s u s e e g n i i F M l
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Jones (1995) presented a radical alternative hypothesis for the structure of the EC. He proposed that both the EC and the Sierra de Perijá ( Fig. 5.3) f orm elements of a single southeast-verging thrust sheet with minor imbrications. A single b asement-involved fault-bend fold is interpreted to underlie the whole mountain range in this model, with a huge triangle zone at the eastern front. The west-verging structures in the Magdalena Valley foothills are interpreted as large gravitational slides developed over a west-dipping ramp. The total shortening according to this model is 185 km. Linares (1996) based on surface geology and limited seismic data interpreted the structure of the EC as the result of inversion of an extensional Mesozoic basin. Reactivation of basement normal faults resulted in both thick and thin-skinned structures related to a single tectonic event. Linares (1996) estimated a minimum of 50 km of shortening at a regional cross-section through the southernmost part of the Sabana de Bogotá area (20 % for the whole Cordillera). A similar interpretation was proposed by Fajardo-Peña (1998) for the Tunja axial region of the EC. However, he proposes five cycles of accumulation and deformation between late Maastrichtian and Holocene times with three peaks of deformation during the Early-Middle Eocene, the Early-Middle Miocene and Late Miocene-Early Pliocene. Inversion of the Magdalena-Tablazo graben (between the La Salina-Landázuri and Boyacá palaeonormal faults) and the Cocuy half graben (between the Soapaga Fault and the Guaicáramo palaeonormal fault) generated the western and eastern flanks of the EC. The Floresta less subsiding block during Mesozoic time correlates with the Tunja axial region where Neogene sedimentary record has been preserved. Tertiary deformation began by inversion of the MagdalenaTablazo graben producing folding within a regional pop-up structure. The Soapaga Fault is interpreted as a footwall shortcut of the Boyacá Fault (Fajardo-Peña, 1998). This author recognized thin-skinned thrusting in the Tunja and Sogamoso regions with detachments in Lower and Upper cretaceous shales and possible positive flower structures in the northern part of the Sogamoso region. Based on seismic data in the Middle Magdalena Valley foothills , Kovas et al. (1982) suggested that hanging-wall anticlines (Payóa-Provincia and La Salina oil fields; Fig. 5.3) were formed above west-directed en échelon thrust faults on the western side of the Nuev o Mundo Syncline (Fig. 5.3). They suggested that the thrust faults are probably flexural-slip features associated with th e folding of the syncline. They also showed progressive transfer of displacement from the Payóa to th e La Salina faults ( Fig. 5.3) arranged in échelon. In the hanging-walls of these faults they als o described backthrusts. To the southward, Steuer et al . (1997) described the Opón structure (gas field, Fig. 5.3 for location) as a large antiform located along the La Salina Fa ult (Fig. 5.3), comprising a wedge bounded by upper and lower detachments with opposite vergence. Th ey interpreted it as the result of a complex history with three episodes: (1) Post-Paleocene to pre-late Eocene compressional deformation which exhumated and eroded the marine Cretaceous rocks; (2) Deposition of the Upper Eocene to Middle Miocene dominantly fluvial detrital rocks; (3) Folding and thrusting beginning at approximately 10.5 Ma ago. Even further southward in the Magdalena Valley foothills area and western flank of the Cordillera, Restrepo-Pace et al . (1999a,b) interpreted 3 events in the west-verging fold-and-thrust belt, which consists of a complex thin-skinned thrust system resulted from polyphase deformation: (1) Westverging thin-skinned closely spaced imbricate thrusts involving Paleocene sediments, concealed below the Eocene unconformity; (2) Reactivation of these structures in a break-back sequence that began in mid-Late Miocene time as indicated by syn-kinematic deposits and onlap relations. That event generated frontal intercutaneous wedges and hinterland west-verging closer spaced thrusts; (3) Major inversion of fundamental hinterland faults creating the present structural relief. In the Llanos foothills, Rathke and Coral (1997) described the Cupiagua condensate- gas field (Fig. 5.3) as a tight asymmetric thrust-controlled anticline with a frontal very steep to overturned and highly imbricated limb. The back limb is broken by several backthrusts. A major tear fault associated with a lateral thrust ramp crosses the structure from west to east. They interpreted that the early development of the structure was simultaneous with deposition of the lower part of Carbonera Fm (Oligocene). De Freitas et al. (1997), Branquet et al. (1999b) and Branquet (1999) have emphasized the importance of strike-slip faults in the EC. Branquet (1999) and Branquet et al. (1999b) identified enechelon folds in the hanging-wall of major basement-involved faults, while basement pop-up structures
188
Chapter 5
limited to the east by convex-upward reverse faults were interpreted by them as positive flower structures. They proposed that these structures formed as the result of dextral transpressional inversion of an Early Cretaceous half-graben during the Andean orogeny. They also emphasize that Andean inversion of normal faults would be impossible without involving an important strike-slip component. Based on outcrop structural data Kammer (1993b, 1996) proposed the existence of hig h-angle normal and reverse faults affecting the basement of the EC ( e.g. Santander and Floresta Massi fs, Fig. 2.1). According to this author the Santander Massif is a broad basement uplift limited on both fla nks by major reverse faults. He also recognized reverse faults in the eastern border of the Floresta Massif. He described more internal faults as high-angle normal faults limiting half-grabens. He interpreted the Neogene fault pattern of the Santander Massif as inherited from a Jurassic extensional event. In contrast Kammer and Mojica (1995, 1996) recognized that deformation of the sedimentary cover is dominated by folding, implying significant shortening, and interpreted them as having been produced by gravitational sliding of the sedimentary cover detached from the basement. For the folded Cundinamarca sub-basin area, Kammer (1999a) suggests that there is a lack of a deformational contrast between sedimentary cover and basement. Mora and Kammer (1999) have presented an illustrative example of the basement-sedimentary cover relationships in the eastern flank of the EC between the Quetame Massif and the Sabana de Bogotá area. In the Quetame Massif (Chingaza dome) deformation affects both basement and cover and precludes, by its homogeneous nature, a detachment within the Lower Cretaceous shales. The Chingaza dome displays open second-order folds, associated to an axial plane cleavage with pytgmatic folds indicative of local shortening values up to 40%. In contrast the Sabana de Bogotá area folding occurred by flexural slip, layer parallel strain is absent, and geometry of the folds suggests detachment horizons within the sedimentary cover. In the western flank of the EC west of Bogota similar observations have been described by Cardozo-Puentes (1989). Based on characteristic relay patterns of marginal folds at both borders of the EC: right-stepped fold pairs along NW- trending borders and left stepped structural relays along N-S trending deformation fronts Kammer (1999b) suggested a transpressional deformation with a strain direction of about 300°. He compared these patterns with outcrop fault slickenside measurements mostly indicative of strike-slip deformations, their populations define, pseudo-conjugate sets, whose compression or extension directions are indicated by reverse or normal fault pairs. In the Santander Massif compression directions are oriented both perpendicular to the NNW striking Bucaramanga Fault and to N-S trending, faults. To the west of the Bucaramanga Fault and affecting the EC the striated faults suggest a clockwise departure from the perpendicular to the structural trend. In contrast, joints evidence a prefolding origin. This author proposed that the model of simple transpression apply to the EC while in the Santander Massif simple transpression is partitioned into a strike-slip deformation along the Bucaramanga Fault and a homogeneous pure shear in the internal part of the massif. Casero et al . (1995, 1997) proposed that the structure of the EC is the result of reactivation of previous structures. They suggested several tectonic episodes to explain the build up of the EC: (1) Strong compressive deformation of Lower Palaeozoic (Lower Ordovician) rocks and subsequent peneplanation; (2) Late Cretaceous flexuration followed by a strong orogenic deformation predating deposition of Paleocene regressive sands; (3) Late Eocene compressional deformation and subsequent moderate erosion; (4) Oligocene-Early Miocene flexuration of the EC foreland followed by moderate compressional deformation; (5) Main Andean orogeny until Miocene-Pliocene times and is still active. They suggest that the EC corresponds to the Lower Cretaceous rift area and represents a megainversion and shortening of the same. Based on surface geology and limited seismic data Schelling (1994) interpreted the structure of the EC as a large basement rooted tectonic pop-up block. They made six regional balanced crosssections across the Cordillera (one in the Upper MV). They proposed that the LLA and Middle Magdalena fold-and-thrust belts ( Fig. 5.3) are divergent, thin-skinned deformational systems, that define the borders of the Cordill era, while the other internal structures are characterized by basementrooted thrust faults and associated basement cored, hanging-wall structural uplifts. They interpreted the Cordillera as a series of inverted graben and half-graben structures. They estimated tectonic shortening between 45 to 65 km, or 18 to 26%.
189
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
4. REGIONAL MAP VIEW RESTORATION OF THE NW CORNER OF SOUTH AMERICA 4.1. METHOD Our aim is to restore the Andean deformation of the EC. In order to establish some boundary conditions for the restoration, I first carried out a regional map view restoration of the NW corner of South America, which defines the tectonic framework for our more local restoration of the EC. I applied a methodology similar to that of Laubsher (1987). I have manually restored in map view a mosaic of regional fault-bounded blocks using all the kinematic constraints available in literature. I followed these steps: (1)
(2)
(3) (4)
I compiled a regional tectonic map, including tectonic plates, deformation belts in the North Andes and major structures ( Fig. 5.1). Between the major tectonic plates, i.e. South America, Nazca and Caribbean, I identified the northern Andes and the Panama Block as deformed belts. The north Andes deformed belt consists of (1) a deformed margin of the South American Plate bounded to the west by the Romer al suture zone and (2) the accreted oceanic terranes of western Colombia and northern Venezuela (Fig. 5.2). I selected a number of tectonic blocks in the Eastern Andes. In order to simplify the model I have used the minimum number of blocks ( Fig. 5.5). Because I am interested in restoring the deformation of the EC (including the Santander Massif) and its two northern prolongations, the Sierra de Perijá and the Mérida Andes, I assume them to be deformed belts located between rigid blocks ( Fig. 5.5). The stable South American Plate is the larger stationary reference block ( Fig. 5.5) and displacements of all other blocks are related to it. To restore the deformation of the deformed belts (EC, Sierra de Perijá and Mérida Andes) I assumed the Maracaibo, C entral Colombia (Centr al Cordillera and Magdalena Valley) and Santa Marta blocks to be rigid ( Fig. 5.5). This assum ption is based on the observation that during Neogene time their in ternal deformation is small, most deformation in these areas appears to have occurred durin g the Palaeogene. I compiled all the kinematic constraints available in literature such as strik e-slip displacement estimates and amounts of shortening from available balanced cross-sections ( Figs. 5.5 an d 5.6, Table 5.1). I manually restored the mosaic of major tectonic blocks according to the kinematic constraints (Fig. 5.7) by means of rigid body translations and rotations. This regional restoration is similar to that done by Laubsher (1987) and by Villamil and Pindell (1998) in northern South America, although their work did not apply to the EC.
A: BALANCED CROSS-SECTIONS USED IN THE MAP VIEW RESTORATION No. Reference 1 to 4 5 6, 7 8 9 10 11 12 13 14 to 16 17, 18 19 20
Findlay (1988) Roeder and Chamberlain (1995) Colletta et al.( 1997) Kellogg (1984) Namson et al. (1994); Linares (1996) Amézquita and Montes (1994) Amaya and Santamaría (1994) Etayo-Serna and Florez (1994) Laubscher (1990); Dengo and Covey (1993) Covey and Dengo (1987); Dengo and Covey (1993) Toro (1998) Laubscher (1990); Namson et al.( 1994) Butler (1983); Ressetar and Schamel (1986); Butler and Schamel (1988); ESRI-HOCOL (1989); Laubscher (1990)
Table 5.1 A and B: Structural cross-sections used in the map view restoration of the Eastern Cordillera. Location of the sections is shown in Fig. 5.6.
190
Chapter 5 No. 21 22 23 to 25 26, 27 28 29 to 31 32 33 34, 35 36 to 41 42 to 45 46 47, 48 49 50 to 64 65 to 70
Reference Kellogg and Duque (1994); Kellogg et al.( 1995) Dengo and Covey (1993) Chigne and Rojas (1997) Cooper et al.( 1995) Ecopetrol et al.( 1995) Naar and Coral (1993) Colletta et al.( 1990) Butler and Schamel (1988) Butler (1983); Ressetar and Schamel (1986); Butler and Schamel (1988); ESRI-HOCOL (1989) Schelling (1994) Covey and Dengo (1987) Laubscher (1990); Galindo et al.( 1994) Galindo et al.( 1994) Geotec (1994) Peel and Hossack (1987) British Petroleum (1989)
B: NOT BALANCED CROSS-SECTION WITH APPROXIMATE BED LENGTH PRESERVATION, USED IN THE MAP VIEW RESTORATION. No. Reference 71 Casero et al. (1995, 1997) 72, 73 Linares (1996) 74 Ecopetrol et al. (1994) 75 Steuer et al. (1997) 76 Ecopetrol et al.( 1995); Ecopetrol-ICP (1995) 77 Ecopetrol-ICP (1995) 78, 79 Butler (1983); Ressetar and Schamel (1986); Butler and Schamel (1988); ESRI-HOCOL (1989) 80 to 82 Ecopetrol Informe No 2133 Tunja area 83 to 102 Ecopetrol and Beicip (1995) 102, 103 Ecopetrol and Beicip (1995); Linares (1996) 104, 105 Ecopetrol and Beicip (1995) 106 Casero et al. (1995, 1997) 107 Corredor (1997) 108 British Petroleum (1987, 1988, 1989) 109 Butler and Schamel (1988) 110, 111 Laubscher (1990) Table 5.1 A and B: Continued
4.2. RESULTS According to the regional map restoration of NW South Amer ica (Fig. 5.7), convergence and clockwise rotation of the Central Colombia micro-plate (Central Cordillera) relative to South America resulted in transpresion with northward increasing shortening in the EC of Colombia. In Ecuador the Andean Block was mainly displaced by right-lateral strike-slip (50 km) with only minor shortening. In contrast the maximum shortening of the EC (from 324 km to 210 km, i.e. 114 km) and clock-wise rotation (+6.5°) of the Central Colombia Block (Central Cordillera) occurred in the north (north of latitude 6° N) at the latitude of the Chocó-Panamá Block. Although the restoration does not explain the cause of Andean deformation, this result is in accordance with the hypothesis that the Andean deformation resulted from the collision of the Chocó-Panamá Block during Neogene time as suggested in literature ( e.g. Duque-Caro, 1990). However, for the EC the maximum amount of shortening (from 324 km to 210 km, i.e. 114 km) occurred between 6 and 7° N, related to the inversion and thrusting of the Mesozoic Cocuy and Tablazo sub-basins. Although the amount of E-W shortening between the Central Colombia Block and the South American increases northwar d, it has partially been absorbed by the right-lateral strike-slip displacement of the Boconó Fault (60 km ; Fig. 5.7). The restoration
191
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera C. COMPLEMENTARY NOT BALANCED STRUCTURAL SECTIONS USED TO CHECK CONSISTENCY OF THE MAP VIEW RESTORATION. NORTHERN VENEZUELA LOWER MAGDALENA VALLEY SANTANDER MASSIF Avé Lallemant (1997) Breen (1989) Boinet (1985) Duval and Valdés (1995) Duque-Caro (1979, 1984) Boinet et al. (1985) Kammer (1993b) MARACAIBO BASIN MIDDLE MAGDALENA VALLEY Colletta et al.(1997) Ecopetrol et al. (1994) Laubscher (1987) Laubscher (1987) Kovas et al. (1982) EASTERN CORDILLERA Roure et al.(1997) Mojica and Franco (1990) British Petroleum (1987-1989) Namson et al.(1994) Calvache and Muñoz (1984) SIERRA DE PERIJÁ Araujo (1997) ESRI - HOCOL (1989) Cardozo-Puentes (1989) Kellogg (1984) Campbell and Burgl (1965) UPPER MAGDALENA VALLEY Laubscher (1987) Butler and Schamel (1988) Casero et al. (1995, 1997) Testamarck et al. (1994) Galindo et al.(1994) Fabre (1987) Mojica and Franco (1990) Fajardo and Chamorro (1994) MÉRIDA ANDES Laubscher (1987) Ojeda and Peña (1994) Irving (1971) Meier et al.(1987) Ressetar and Schamel (1986) Julivert (1970) Toro (1992) Kammer (1993b) LLANOS ORIENTALES Ecopetrol and Beicip (1995) Van der Wiel (1991) Kammer (1999a) Kluth et al. (1997) GARZON MASSIF AND FOOTHILLS Kammer and Mojica (1995) McCollough and Carver (1992) Butler and Schamel (1988) Kammer and Mojica (1996) Casero et al. (1995, 1997) Keeley and Súarez (1997) LLANOS FOOTHILLS British Petroleum (1987-1989) Kammer and Mojica (1995) Laubscher (1990) Casero et al.(1995, 1997) Laubscher (1990) Linares (1996) Ecopetrol-ICP (1995) Van der Wiel (1991) McLaughlin (1972) Linares (1996) Occidental et al. (1996a,b) PUTUMAYO-ORIENTE BASIN Naar and Coral (1993) Baby et al.(1998) Toro (1990) Rathke and Coral (1997) Dashwood and Abbotts (1990) Torres (1996) Ureta and Du Toit (1997) Ujueta (1993) Table 5.1 C: Continued
(compare Figs. 5.5 a nd 5.7) suggests that the total amount of shortening perpendicular to the deformed belts in the north (1 0 km of sh ortening has been estimated in the Sierra de Perijá and 50 to 60 km of shortening in the Mérida Andes) is less than the maximum shortening (114 km) of the EC between 6 and 7° N. Convergence (from 100 to 65 km, i.e. 35 km) and left-lateral strike-slip movement (100 km) of Central Colombia (Central Cordillera and MV) relative to Maracaibo produced the northern part of the EC in the Santander Massif r egion. Clock-wise rotation of the Central Colombia Block relative to Maracaibo is +3.5° ( Figs. 5.5 a nd 5.7). Convergence (from 45 km to 35 km, i.e. 10 km) of the Santa Marta Block relative to Maracaibo generated the Perijá Mountain Range. Restoration suggests also that clock-wise rotation of the Santa Marta Block relative to the Maracaibo Block (+7°) took place. Finally convergence (50 to 60 km) and dextral strike-slip motion (60-km) of t he Maracai bo Block relative to South America generated the Mérida Andes of Venezuela (Figs. 5.5 a nd 5.7).
5. MAP VIEW RESTORATION OF THE EASTERN CORDILLERA 5.1. METHOD I carried out a more detailed map view restoration of the EC of Colombia (Figs. 5.1, 5.2 and 5.3), using balanced cross-sections from the literature ( Figs. 5.4, 5.6 a nd Table 5.1). The aim is to reconstruct the undeformed initial state. I used a modif ied version of the method applied by Bourgeois et al . (1997) to restore thrust sheets in map view in the Tajik Depression, central Asia.
192
Chapter 5 15 N 14 N
13 N
CARIBBEAN PLATE
GUAJIRA BOCK
12 N
11 N
SANTA MARTA B BLOCK
10 N
R i n a t n o u M a MARACAIBO j i r BLOCK P e
u c a r a m S a n a n g ta a F n d a e u r l t M a s s i f
9N
8N
7N
k c lo B ia b o m l o C
6N
NAZCA PLATE 5N
4N
S E N A R R E T D E T E R R C A C I N E A C O E O A L P A
3N
2N
1N
0
1S
e s n d A i d a r e M
r a l e l i d o r C r n t e s E a
l r a t n e C
l t a u F i r a m a l t A
Detail in Fig. 5.8
SOUTH AMERICAN PLATE Continental plate and tectonic blocks
Deformed belts Kinematic constraints and location of selected balanced cross sections from literature
2S
3S
Shortening Strike-slip displacement
4S 5S 83 W
0 82 W
81 W
80 W
79 W
78 W
77 W
76 W
75 W
74 W
73 W
72 W
500 km
71 W
Figure 5.5. Tectonic blocks used in the regional map view restoration of the NW corner of South America. Kinematic constraints include: (1) Regional balanced cross-sections across the Sierra de Perijá (Kellogg, 1984), the Mérida Andes (Colletta et al., 1997), and the EC (Colletta et al., 1990; Dengo and Covey, 1993; Schelling, 1994; Cooper et al., 1995; Roeder and Chamberlain, 1995; Linares, 1996); (2) Strike-slip displacement estimates along the El Pilar-Oca fault system (Pindell, 1993; Villamil and Pindell, 1998), Boconó Fault (Schubert, 1982; Henneberg, 1983; Pindell, 1993; Pindell and Erikson, 1993; Colletta et al., 1997), Bucramanga-SantaMarta Fault (Campbell, 1968; Irving, 1971) and Altamira Fault (Casero et al., 1995, 1997). In the restoration I assumed the South American, Central Colombia, Maracaibo, Santa Marta and Guajira blocks were rigid and their relative Neogene movement generated deformed belts of the EC (including the Santander Massif) and its northern prolongations the Sierra de Perijá and the Mérida Andes. I did not include in the restoration oceanic accreted terranes, the Chocó-Panamá Block or the Nazca and Caribbean plates.
193
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
0
8
100 km
B u c a r a m a n g a fa u l t
a s u c L n a S d e i a n a r r e S
7 S ie r r a d e P e r i j a
i d a r M e
S a n ta Cucuta n d e r M a s s i f Bucaramanga
s d e A n
4 3
2 6
Barinas
1 Arauca
88 52
Medellin
25 83
89
ra i le r d o C
Llanos Orientales Basin (LLA)
30 Tunja Yopal
Manizales
19 22 Bogota
Ibague
21 85 Villavicencio
47
l a t r n e C
l t F a u a r i a m A l t 35 Neiva
e a d n e a r i n a c a a r r M e a S L
71
i f s s a M o n z r G a
Structural cross-sections used in the map view restoration 6
Structural cross-sections where bed shortening was measured (see Table 5.1 A and B). Complementary not balanced structural sections used to check consistency of the map view restoration (see Table 5.1 C).
Guyana Shield
Figure 5.6: Location of the structural cross-sections used in the map view restoration of the Eastern Cordillera. Numbers refer to Table 5.1.
194
Chapter 5
0
500 km GUAJIRA BLOCK
B u c a ra m K a C n O g a L B F a u A I l t B M O L O C L A R T N E C
SANTA MARTA BLOCK
a r ij P e d e
MARACAIBO BLOCK
s d e n A a u l t a F r i d n o e o c M B o
i a a n r e r S
9N 8N
7N
r a l e l i d r o C
6N
5N
n r e t s a E
4N 3N
l t u F a i r a m a t l A
2N
Detail in Fig. 5.10 1N
SOUTH AMERICAN PLATE
0
1S Continental plate and tectonic blocks Deformed belts Kinematic constraints and location of selected balanced cross sections from literature 79 W
78 W
77 W
76 W
75 W
74 W
73 W
72 W
71 W
70 W
Figure 5.7. Palaeogene restoration of the major tectonic blocks of the NW corner of South America before the Neogene Andean deformation. The co-ordinate grid represents the present day geographic grid restored to its location before Neogene deformation.
They reconstructed the undeformed state of a single stratigraphic layer, currently folded and offset by reverse faults. First they represent in map view the present, deformed state of the stratigraphic surface to be restored as a mosaic of folded blocks bounded by faults. On this mosaic, blocks may be contiguous, separated by gaps, or overlapping each other, according to the nature of the faults (strikeslip, normal or reverse, respectively). The width of the gap or overlap between blocks is proportional to the heave of the associated fault. Second, they unfold the blocks. Unfolding changes the shape of each block and the width of overlaps at their boundaries. Third, they packed the unfolded blocks using rigid translations and rotations to minimise the total area of gaps and overlaps. This yields a restored mosaic. Because I lack structure contour maps at particular stratigraphic horizons in the EC, their method cannot be applied directly.
195
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Additionally, in the EC the erosion level is highly variable with the result that any particular stratigraphic horizon may lie deep (without ou tcrop control) in areas of shallow erosion level or may be completely eroded in areas of deep erosion (Fig. 5.9). Clearly in these cases there is no structural control. Structural control for most of the sections is limited to surface geological maps and outcrop data. Bed length preservation can only be applied to deep stratigraphic horizons in areas of deep erosion level and to shallow stratigraphic horizons in areas of shallow erosion. In addition the location of faults is known only from surface geological maps. In order to deal with these limitations, instead of measuring bed lengths at a particular stratigraphic horizon, I measured both the undeformed and deformed bed lengths, for each block, referred to the present-day topography surface and its restored geometry before deformation (Fig. 5.9, see explanation below). I used the following procedure: (1) I compiled a detailed structural map of the EC ( Fig. 5.3). (2) I compiled as many as possible bala nced structural cross-sections of the EC from available literature (Fig. 5.6 and Table 5.1) and I checked their consistency in order to discard a few number of clearly inconsistent sections ( c.f. Roeder and Chamberlain, 1995 cross-section; see discussion in section 6.2 of this chapter). (3) I defined a number of blocks limited by faults (Fig. 5.8). I assumed that it is reasonable to divide the region to be restored into a mosaic of blocks completely bounded by faults. Since natural faults arrays are seldom totally connected, the mosaic has to be completed by introducing artificial block boundaries. I extrapolated each fault trace until it meets another fault trace. To allow some bending within blocks that are far from equant, I subdivided them into smaller ones, using new artificial block boundaries. These artificial block boundaries were drawn on the basis of shape of the blocks and where possible following minor faults or tight folds. Blocks so defined are assumed to consist of folded layers. This assumption is reasonable if surface strains are negligible. (4) For each block I measured along each balanced cro ss-section the present-day deformed length and the undeformed (pre-Andean) restored length ( Figs. 5.4 an d 5.9). I also measured the horizontal displacement for each fault bounding a blo ck (Figs. 5.4 a nd 5.9). Because there are no accurate structural contour maps of particular stratigraphic reference levels, and the location of faults is only known from surface geological maps, I measured both the undeformed and deformed bed lengths, for each block, as the bed length between the location of the surface trace of the faults limiting the block (Figs. 5.4 and 5.9). In the restored undeformed state I also measured the horizontal separation of the p resent-day topographic surface, and its restored geometry before deformation, across each boundary fault, and I distributed this length (L f , see Fig. 5.9) in two equal parts that were added to the restored length of the two blocks limited by the fault (Figs. 5.4 an d 5.9). In this way the restoration was done not at a particular stratigraphic surface but at the pr esent-day topographic surface. Although this method can produce small differences as compared to a restoration at a particular stratigraphic horizon, the total amount of shortening between pin lines is the same. Restorations made along different stratigraphic horizons can differ but the total amount of shor tening between pin lines will remain essentially the same (Fig. 5.9). (5) I unfolded each block separately ( Fig. 5.9). I assumed that the stratigraphic horizons before deformation were planar. This assumption is reasonable, as sedimentary irregularities are small in comparison with the horizontal scale of the restoration. Obviously this restoration takes into account all the deformation since deposition. In general, for non-cylindrical folds, it is convenient to use automatic unfolding methods (Gratier et al ., 1991). For the EC, where folds are nearly cylindrical, I have performed this task manually. On a series of cross-sections, drawn perpendicular to the main structures, I have conserved bed lengths. This process yielded an extension of the blocks in map view parallel to the directions of the cross-sections. After each block has been unfolded, it was assumed to be rigid.
196
Chapter 5
(6)
(7)
I manually restored the relative location of each fault-bounded block to the pre-deformation state fitting in map view the original undeformed lengths and minimising the distances across fault separations ( F igs. 5.9 and 5.10). I manually packed the blocks using rigid body translations and rotations, visually minimising the total area of gaps or overlaps. This yielded a restored block mosaic. Boundary conditions on displacements were conveniently introduced by keeping one block stationary. I choosed the cratonic stable South American Plate as the stationary block: All the displacements were referred to this. By comparing the material present-day co-ordinate grid and its location in the restored predeformed (pre-Andean) state, I obtained an approximate field of finite horizontal displacements (Fig. 5.11).
5.2. RESULTS Figure 5.10 shows the map view restoration of the EC. The overall fit of the restored fault block map is visually satisfactory. According to the restoration the amount of shortening of the EC is approximately one half of the present-day width of the Cordillera. Both width and amount of shortening perpendicular to the mountain range increase northward ( Table 5.2). This is the result of clock-wise rotation (+6°) of the Central Colombia Block (Central Cordillera and Magdalena valley) relative to the stable South American Plate. In addition to shortening normal to the axis of the Cordillera there are important strikeslip components along the Bucaramanga Fault (left-lateral 100 km) in the north and the Altamira Fault (right-lateral 30 km; Fig. 5.10). ORIGINAL DEFORMED SHORTENING LOCATION LENGTH L0 LENGTH L0-L1 (KM) (KM) L1 (KM) Northern Magdalena-Tablazo and Cocuy sub-basins 360 240 120 Southern Magdalena-Tablazo and Cocuy sub-basins 340 230 110 Northern Cundinamarca sub-basin 300 210 90 Southern Cundinamarca sub-basin 200 140 60 Neiva sub-basin (Upper Magdalena Valley100 90 10 Southern EC)
L1/LO (%) 66.67 67.65 70.00 70.00 90.00
Table 5.2. Amount of shortening obtained fr om the map view restoration along several cross-sections perpendicular to the EC. See Figures 2.1 and 5.8 for location of regions mentioned.
The Upper Magdalena Valley and south-eastern part of the Cordillera were displaced transpressionally toward the NE with respect to the stable Llanos area (South American Plate) as proposed by Branquet (1999) and Branquet et al. (1999b). Left-lateral displacement of the Bucaramanga-Santa Marta Fault is accommodated by shortening and r elative left-lateral displacement of blocks in the western flank of the EC (Magdalena-Tablazo sub-bas in, Fig. 5.10). Space constraints from the restoration also suggest N-S shor tening in the western flank of the Cordillera. In this restoration I assumed that the Garzón Massif (Fig. 5.10) was displaced northeastward relative to stable South American Plate, as suggested by Casero et al . (1995, 1997). The Serranía de La Macarena Block (Fig. 5.10) represents the leading edge of the Garzón Massif systems of blocks. Results of th e map view restoration suggest that the SE flank of the EC (Cocuy and eastern Cundinamarca sub-basin , Fig. 5.10) was right-lateral transpressively deformed, but that the NW flank of the EC (Magdalena-Ta blazo sub-basin, Fig. 5.10) was left-lateral transpressively deformed.
197
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera 1700
0
100 km B u c a ra m a n g a F a u l t
1600
1500
K C O L s B a c IA L u B n M S a O L d e O a C a n i L rr A S e R T N E C
1400
1300
1200 r a il e rd o C
Manizales
1000
Ibague
90 0 l a tr n e C
80 0
70 0
a
i r n a l t a m e l d a A a g M y r l e p a l e U p V
Neiva
n r z o G a
800
Maracaibo Basin
s d e n A a d i r M e
Serrania de Perija
Bucaramanga
S e r r a n i a M a s s i f
Cucuta
Barinas basin Arauca
i n a s b b S u o a z l b T a u y a Tunja o c i n n C s l e a a d d t e b b r a g v e u M I n S d t e e r v I n i s n Bogota B a c a a r m n a i n d C u Villavicencio r n t e s E a
Medellin
1100
ri ja e P e d a r r ie S
Yopal
Continental plate and tectonic blocks
u l t F a
Deformed belts
a e n e d r a i a n c a a r r M e a S L
Kinematic constraints and location of balanced cross sections fron literature
f s s i M a
Colletta et al. (1990) section (Figure 5.4)
Guyana Shield
90 0
Llanos Orientales Basin (LLA)
1000
1100
1200
1300
1400
Figure 5.8. Present day mosaic of fault limited blocks used in the map view restoration of the EC and location of balanced cross-sections from literature used in the restoration. Black line represents the balanced cross-section by Colletta et al . (1990) shown in Figure 5.4.
198
Chapter 5
Fault bounded block section length after deformation WNW n i P a n e l a d g a M
t s u r h T a n i l a S a L
t s u r h T a r r a t i m i C Tm
Tm Tr
Tr Tchu
T r T c h o c h u T
Tchu
Tcho
t s u r h T s a m r A Kl
t s u r h T a t r o H
Kpts Kpts
Ku Kl
ESE
Kpts
Ktr
Jg
Kpts
Kpts
K u K l
Tcho
t s u r h T i r u z a d n a L
n i P e n i l c i t n A s e n o t r o P
Jg Ktr
Ku
Jg
Jg
Ktr Jg
Kpts
B
Ktr Jg
B
B B
B
B
Fault bounded block section length before deformation n i P a n e l a d g a M
Tm Tm Tr Tchu
L f
t s u r h T a r r a t i m i C
t s u r h T a n i l a S a L
2
Lf
t s u r h T a t r o H
Tr Tchu Tcho
Tcho Ku
Ku
Ku Kl
LEGEND Tm
t s u r h T i r u z a d n a L
t s u r h T s a m r A
n i P e n i l c i t n A s e n o t r o P
Kl
Kpts B
Ktr
Chorro Gp.
Ku
Umir Fm.
Kl
La Luna Fm.
Kpts
Paja, Tablazo and Simiti Fms.
Ktr Jg
Tambor and Rosablanca Fms. Giron Gp.
B
Basament
Ktr
Jg
Jg
Ktr Jg
Jg
B
Tr Real Fm. Tchu Chuspas Gp. Tcho
Kpts
Kpts
Mesa Fm.
Ku
B B
10 km
B
B
Horizontal and vertical scale
Figure 5.9. Method used in the map view restoration of the EC. This example is a fra gment of the western end of the structural cross-section BB’ from Schelling (1994). Location of this section is shown in figure 5.8.
199
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
1 6 0 0
S a n t a M a r t a - B u c a r a m a n g a F a u lt
1500
1 4 0 0
K C L O B 1 30 0 IA B M O y l e L O V a C a L 1 2 0 0 A e n l R d a T g a N M E n a e l C l e d a d a g i d M M d 11 0 0 t e
a j i r e P e d a r r e i S
b u b S o l az b T a
a e r d i l o r C r n s t e a E
i n s a b b u S
l s l u y i c h t o C f o o d t e o s r n e v l a L I n i n n b a s r t e b a s S u E d a t e a r c r e v m I n i n a d u n C
u e g a b I
90 0
l t u F a
n a e l d a y g a l e M l a r V e p p U
0
a e n d e r a a i n c a a M r r e a S L
80 0
70 0
n z o a G
i f s s a M 800
a r i d e M
s d e n A
S e r r a n i a M a s s s in i f a
r v e I n
1 0 0 0
o n b i s m a u b t a b t a u C S
90 0
100 km
Continental plate and tectocnic block Deformed belts Kinematic constrains and location of balanced cross sections from literature.
1000
Figure 5.10. Palaeogene restoration of the mosaic of fault limited blocks before Neogene deformation. Grid represents the present day plane co-ordinate grid restored to its location befor e Neogene deformation. Thick black broken line represents the Colletta et al . (1990) balanced cross-section shown i n figure 5.4.
200
Chapter 5
5.3. DISPLACEMENT OF BLOCKS RELATIVE TO STABLE SOUTH AMERICA AND ROTATIONS ABOUT VERTICAL AXES By comparing the material present-day co-ordinate grid and its location in the restored p reAndean state, I obtained an approximate field of finite horizontal displacements ( Fig. 5.11). The displacement vectors increase in length from east to west, because they are refer red to the eastern cratonic South American Block, which is stationary. Several regions can be distinguished according to their displacement vectors ( Fig. 5.11) (1)
(2)
(3)
(4)
(5)
The Garzón Massif displacement vectors are toward NE and their magnitude decreases from about 30 km in the SW to almost zero in the Serranía de la Macarena. Rotation of these blocks about a vertical axis is practically negligible. This region is separated from the southern EC and Upper Magdalena Valley by the Altamira Fault. The Upper Magdalena Valley-southeastern flank of the EC (eastern Cundinamarca sub-basin and Cocuy sub-basin) and southern Central Cordillera (south of the Ibagué Fault) displacement vectors are toward NNE in the Upper Magdalena Valley and gradually change toward E in the Cocuy area. Their magnitude gradually increases NW to a maximum of 90 km in the Girardot sub-basin of the Upper Magdalena Valley. Blocks in this region show clockwise rotation about a vertical axis. These rotations decrease toward NE to a minimum in the Cocuy area. The western flank of the EC (Magdalena Tablazo sub-basin and western part of Cundinamarca sub-basin) and northern part of Central Cordillera (north of the Ibagué Fault) displacement vector azimuths are ESE but gradually change toward SE in the Magdalena-Tablazo sub-basin close to the Bucaramanga Fault. Displacements in this region increase northward as this region has rotated in a clockwise sense about a vertical axis close to its southern tip. Displacement also increases westward. Between the regions 2 and 3 I assume a left-lateral shear zone. The Santander Massif–Sierra de Perijá region displacement vectors are toward E in its southern part, but change to ENE in its northern part. The displacement magnitude increases northward as this region has been rotated over a small angle about a pole close to its southern tip. Strong changes in the displacement field between regions 3 and 4 are accommodated by the mostly leftlateral strike-slip relative displacement of the Santa Marta-Bucaramanga Fault. The Mérida Andes-Maracaibo Basin has been displaced toward the east. Displacement increased westward within the Mérida Andes. Rotation of blocks about a vertical ax is negligible in this region. A complex zone is located between the Cocuy sub-basin, the Santander Massif and the Mérida Andes.
6. DISCUSION 6.1.
ADVANTAGES OF THE MAP VIEW RESTORATION
As discussed in the results, the method I applied is useful to constrain shortening and strike-slip displacement estimates as well as to detect strike-slip motions and rotations about vertical axes not revealed by balancing cross-sections. Obviously results should be checked with other independent methods.
6.2.
COMPARISON WITH PREVIOUS SHORTENING ESTIMATES
Table 5.3 s hows a comparison between some shortening estimates from literature and those obtained in the map view restoration. The amount of shortening perpendicular to the EC obtained in the map view restoration is greater than the shortening estimates from Cooper et al . (1995), Linares (1996) and most of the balanced cross-sections of Schelling (1994). Map view shortening estimates are, however, smaller than those estimates from Colletta et al . (1990), Dengo and Covey (1993) and Roeder and Chamberlain (1995). The smallest difference 9% is for the Colletta et al. (1990) section. The shortening figure of Dengo and Covey (1993) seems to be overestimated. The balanced-cross-section
201
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
1 6 0 0
S i e r r a d e P e r i j a
1500
a r i d e M
1 4 0 0
K C O L B IA B M O L O C L A R T N E C
13 0 0
1 2 0 0
S a n ta n d e r M a s s i f
4
a n l e i n d a a s g b a b M u d s t e z o r e a l v n b I a
3
1 1 0 0
i n s a b b u s y u o c C d t e r e v I n
T
a l e r l i d o r C r n t e s E a
1 0 0 0
900
80 0
u e g a b I
l t u a F
2
y e V a l na a le d a g M p e r p U
1 70 0
a n e e d r a c a r r a e M i S a L
f s s i a M n o r z G a 800
s5 d e n A
900
1000
1100
00
100 Km km 100
1200
1300
Figure 5.11. Displacement field during Neogene deformation of the EC. Each vector connects the present day coordinate grid with its position before Neogene deformation. Numbers represent regions discussed in section 5.3 of this chapter.
202
Chapter 5
interpretation they propose contains a hypothetical low-angle mid crustal detachment that extends westward to the Central Cordillera up to the western edge of the South American plate. According to them, during Miocene to Pliocene shortening in the EC, the Central Cordillera was carried passively up this mid-cr ustal detachment, decapitating and thrusting a Jurassic graben below the Magdalena Valley (Fig. 5.3) and generating regional uplift of the Central Cordillera and erosion of its sedimentary cover. A decapitated graben below the basement rocks of the Magdalena Valley seems to be very speculative and it is not supported by evidence. Roeder and Chamberlain (1995) also overestimated the amount of shortening. Their shortening estimation seems to be unreal. Their cross-section interpretation is based on the questionable assumption that the whole Cretaceous sedimentary record represents a thermal sag basin fill after Jurassic rifting (ignoring earlier work by Fabre, 1987, who demonstrated the existence of an Early Cretaceous extensional basin). Literature Reference
Shortening (km)
This work L1/l0
Shortening (km)
Colletta et al.(1990)
105
66%
95
Dengo and Covey (1993)
150
40%
105
Cooper et al.(1995)
68
80%
85
Roeder and Chamberlain (1995)
230
58%
105
Jones (1995)
185
58%
105
Linares (1996)
50
20%
80
45 to 65
18 to 26 %
50 to 108
Schelling et al.(1997)
Shortening % Differen L1/L0 Difference (km) * #
70.31 % 69.57 % 72.58 % 69.57 % 69.57 % 71.43 % 69.6 to 78.2 %
-10.00
-9.52
-45.00
-30.00
17.00
25.00
-125.00
-54.35
-80.00
-43.24
30.00
60.00
5 to 43
11.11 to 66.15
L0: Restored undeformed length of the section; L1: Present day deformed length of the section * Difference = shortening this work – shortening literature; # % Difference relative to shortening literature as 100% Table 5.3. Comparison of amount of shortening obtained from the map view restoration and some balanced crosssections from literature.
6.3.
COMPARISON WITH OUTCROP STRUCTURAL STUDIES, STRESS INFERRED FORM BOREHOLE BREAKOUT DATA AND PLATE MOTIONS FROM GEOPHYSICAL DATA
6.3.1. Shortenin g per pendi cular to the r egional stru ctur al grai n of the EC
A general shortening perpendicular to the regional structural grain of the EC recognized in the map view restoration (compare Figs. 5.8 a nd 5.10) is supported by structural kinematic fault and fault striae data from the EC ( Figs. 5.12 t o 5.16, Cobbold et al ., 1988; Cardozo and Zúñiga, 1995; Díaz and Sotelo, 1995; Kammer , 1996, 1999 b), by the orientation of systematic conjug ate joints normal to bedding planes interpreted by Kammer (1999b, 2000) as ori ginated before folding (Fig. 5.15) andby the orientation of stylolites and vertical joints of recent origin ( Fig. 5.16, Mojica and Sheidegger, 1981; 1984; Mojica 1983 in Torres, 1992; and Mojica, 1985). Also a general shortening perpendicular to the regional structural grain of the EC is supported, the present stress orientation as inferred from borehole breakout data ( Fig. 5.17, Castillo and Mojica, 1990; Ochóa and Ponguta, 1991 and Torres, 1992) as well as from the orientation of compressional axis of focal mechanisms solutions of earthquakes in the upper 50 km of the crust ( Fig. 5.18, Daniels, 1991). Although fault striae data from the EC approximately suggest shortening perpendicular to the mountain range in detail a clockwise departure from the perpendicular to the structural trend (Fig. 5.14) may be interpreted as a result of a collision of the western block of the Bucaramanga Fau lt with the eastern flank of the EC (Kammer, 1999b).
203
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Mojica and Sheidegger (1981) measured the orientation of recent vertical joints with smooth surfaces in outcro ps (Fig. 5.16). Their data shows large dispersion. They found two sets: N 1°-8° E and N 91°-100° E forming angles close to 90°. The bisectrix of these sets corresponds theoretically to principal stress directions with azimuths N 45 –54 E (approximate parallel to the EC) and N135° – 144° E (approximately perpendicular to the EC). The identification of the maximum and minimum stress is uncertain from the joint measurements alone, their results show consistent orientation o f the principal stress axis, but differ ent identification of the maximum stress axis at neighbour locations ( Fig. 5.16). This inconsistency could be eliminated if all orientations perpendicular to the EC are reinterpreted (c in Fig. 5.16) as those of maximum stress according to inferences from horizontal stylolites and other structural data. Mojica (1983 in Torres, 1992) and Mojica and Scheidegger (1984) studied the orientations of horizontal stylolites, fold axes, deformed fossils, recent joints and fault plane solutions throughout the Colombian Andes and found NW-SE maximum horizontal compression (approximately 140°) with local variations of E-W and WNW-ESE. Mojica (1985) measured tectonic stylolites in the EC between SW of the Neiva City and the southern termination of the Bucaramanga Fault. Stylolites are oriented NW-SE with subordinate NE-SW and E-W directions at few outcrops. These structural data suggest an approximate shortening perpendicular to the structural trend of the EC. From borehole breakout analysis of wells in the main sedimentary basins of Colombia Torres (1992) suggested a pr esent-day stress regime with the maximum horizontal stress oriented WNW-ESE to NW-SE ( Fig. 5.17). She distinguished a northern (above 6° latitude, Guajira, Maracaibo, Catatumbo and Lower and Middle Magdalena basins; Fig. 1.1), central (between 4° and 6° latitude, Llanos and Upper Magdalena basins) and southern (between 0° and 1° latitude, Putumayo Basin; Fig. 1.1) sub-sets with mean maximum horizontal stress orientations of 117°, 111° and 133° respectively. Although this reflects small changes in the horizontal stress orientation with latitude, the horizontal stress orientation trend perpendicular to the EC is regionally preserved. Borehole breakouts indicate a horizontal principal stress of 138° for the Llanos wells and 112° for the Middle Magdalena Valley (Castillo and Mojica, 1990) and NW-SE compression in the Upper Magdalena Valley (Ochóa and Ponguta, 1991, Fig. 5.17). Daniels (1991) compiled the orientat ion of compressional axes from focal mechanism solutions of earthquakes in the upper 50 km of the cru st (Fig. 5.18). His results indicate ENE-SWW compression in the northern part of the EC and E-W to WSW-ENE in the southern part of the EC and Ecuador. As mentioned before, all these data support the shortening approximately perpendicular to the EC suggested by the map view restoration. 6.3.2. Conju gate (?) left-l ater al and ri ght-l ater al str ik e-sli p faul ts
Left- and right-lateral strike-slip fault zones suggested by the map view restoration are supported by outcrop structural data. Left-lateral strike-slip motion of the Buc aramanga Fault ( Fig. 5.3) is supported by kinematic analysis of lineation data on the fault (Figs. 5.12 and 5.13). Acosta et al.( 2000) interpreted from these data that the Bucaramanga and Suarez faults (Fig. 5.3) ar e steep dipping left-lateral strike-slip faults. Two more set of planes were measured in t he Bucaramanga Fault: one set trends sub-parallel to the main structure and is interpreted as riedels. A second set is formed by steep-dipping right-lateral faults; these are interpreted as antithetic riedel shears. A similar pattern was found by Meier et al . (1987) in the Táchira area, which is the link between the EC of Colombia and the Mérida Andes of Venezuela (Fig. 5.3).
204
Chapter 5
Serrania de Perija
100 km
0
Maracaibo Basin
a r i d e M
s a c u L n a S e d ia n a r r e S
Cucuta
7 Barinas
ra l e i rd o C
basin
Bucaramanga
1
a e n l ) d a V a g ( M M y e l l e l d a i d V 2 M
Medellin
4
Arauca
Llanos Orientales Basin (LLA)
3
Tunja
Manizales
5 Bogota Ibague
6
) C E ( A R E L L I D R O C N R E T S A Villavicencio E
Yopal
8 l tr a n e C
s d e A n
y lle a V a Neiva n le a s if d g a a s M M r o n e p z p a r U
G
Guyana Shield
OUTCROP STRUCTURAL DATA (Kinematic indicators) 1. Cobbold et al.(1988)
e a d n e a r a 2. Acosta (2000) i n c a a 3. Kammer (1996); Cardozo and Zuniga (1995) r r M e a S L 4. Diaz and Sotelo (1995)
5. Cardozo-Puentes (1989) 6. Kammer and Mojica (1995) 7. Meier et al. (1987) 8. Kammer (2000)
Figure 5.12. Location of detailed outcrop structural studies of the EC, including kinematic indicators and fault striae from literature used to compare with the results of the map view restoration.
205
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
0
Serrania de Perija
10 0 k m
Maracaibo Basin
s a c u L n a S e d ia n r r a e S
Cucuta B u c a r a m a n g Bucaramanga a F
F s o n a rm e H s o D
Medellin
r a e l l i r d o C
F a d o n H
Manizales
e F b a g u I b
d a r i d e M
s d e n A
Barinas basin Arauc a
F a n F i l z S a a r e u L a S
2 Tunja
4
F a i m t u i B
3 Yopal
Bogota
Ibague
l r a t e n C
y l e a V n a i f a le s s Neiva d a a g M M o n r z p p e a r G U
N E R E R A T S L L E A R D I C O
Villavicencio
Llanos Orientales Basin (LLA)
a e n d e r a a i c n a a r r M e a S L
Guyana Shield
OUTCROP STRUCTURAL DATA 2. Acosta et al.(2000) 3. Kammer ( 1996); Cardozo and Zuniga (1995) 4. Diaz and Sotelo (1995)
| Figure 5.13. Kinematic indicators and other other outcrop structural data in the EC from from literature. literature.
206
Chapter 5
Serrania de Perija
0
100 km
ra i le rd o C
y l e a V a l en a d g a M Bucaramanga l e d i d M
d a r i d e M Cucuta
s d e n A
Barinas basin
Arauca Arauca
A R E L L I D R O Tunja C
Medellin
Manizales
Maracaibo Basin
N R E T S A E
Yopal
Bogota Ibague Villavicencio
l n t ra e C
y a l e V n a le Neiva d a g M i f p e r s s a p n M U o z r G a
LLANOS ORIENTALES BASIN (LLA)
Guyana Shield
a e n d e r a a i c n a a r r M e a S L
OUTCROP STRUCTURAL DATA (Kinematic indicators) indicators) 8. Kammer (2000) Reverse fault Normal fault Strike-slip fault
Figure 5.14. Kinematic indicators from fault striae in the EC. At each locality solid arrows indicate shortening direction, open arrows indicate extension direction and lines intermediate direction. Sterographic plots (equal area lower hemisphere) indicate also the bedding/foliation orientation. Relative fault movement is indicated at the localities where most of measured fault planes with that sense predominate (modified from Kammer, 1999b)
207
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Serrania de Perija
0
100 km
ra i l e rd o C
Medellin
Manizales
Ibague
s a c u L n a S e d ia n y ra r le e a S V a n le d a g a Bucaramanga M l e d i d M
l n t ra e C
d a r i d e M Cucuta
s d e n A
Barinas basin
Arauca Arauca
A R E L L I D R O C Tunja
N R E T S A E
Yopal
Bogota
Villavicencio
y a l e V n s if a le s Neiva a gd M a n e r M z o r p p G a U
Maracaibo Basin
LLANOS ORIENTALES BASIN (LLA)
Guyana Shield
a e n d e r a a i c n a a r r M e a S L
OUTCROP STRUCTURAL DATA (Joint analysis) 8. Kammer (2000) Reverse fault Normal fault Strike-slip fault
Figure 5.15. Inferred shortening and extension directions from joints measured at outcrops. At each locality solid arrows indicate shortening direction, open arrows indicate extension direction and lines intermediate direction (from Kammer, 1999b).
208
Chapter 5 ORIENTATION OF RECENT VERTICAL JOINTS AND INFERRED STRESS ORIENTATION (from Mojica and Sheidegger, 1981)
a a i j i r n e a r P r e e S d
Orientation of vertical joints.
Maracaibo Basin
Inferred compressive stress axis. Average of the region closed by a rectangle. Orientation corrected from the Mojica and Scheidegger c (1981) figure because that maximum stress is inconsistent with horizontal stylolites reported by Mojica and Scheidegger (1984)
s a c u L n a S e d ia n ra r e y l e a S V ra i le rd o C
d a r i d e M
c c
N R E T S A E
Ibague
c
Arauca Arauca
Bucaramanga
c c c
R O C
Manizales
Barinas basin
Cucuta
a e n l d a g a A M R e E l L d L i d I D M
Medellin
c Tunja Yopal
c c Bogota
LLANOS ORIENTALES BASIN (LLA)
Villavicencio
l n t ra e C
y l e l a V n e l d a g c Neiva a M r e i f p s s p U M a c
s d e A n
Guyana Shield
a e n d e r a a i c n a a r r M e a S L
n r z o a G
0
100 km
Figure 5.16. Orientation of recent vertical joints and inferred stress orientation (modified from Mojica and Sheidegger, 1981).
209
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
Maracaibo Basin
Serrania de Perija
e s d A n a i d e r M
s a c u L n a S e d ia n ra r e S
Medellin
r a e l l i d r o C
Cucuta
Barinas basin
Bucaramanga
y l e l a V n a e l d a g a M l e d d i M
Arauca Arauca
A E R L I L D R C O Tunja R N E T S A E
) C E (
LLANOS ORIENTALES BASIN (LLA) Yopal
Manizales
Bogota Ibague
Villavicencio
Guyana Shield
0
l a t r n e C Neiva
PUTUMAYO BASIN
f e a i s d n s e r a M a a i c n a n a o r M z r e a a r G S L
100 km
STRESS FIELD ORIENTATION ( 1 ) INFERRED FROM BOREHOLE BREAKOUT BREAKOUT DATA Decreassing quality rank
D
C
B
A
Torres (1992) Castillo and Mojica (1990) Ochoa and Ponguta (1991)
Figure 5.17. Orientation of present day stress field inferred from borehole breakout studies.
210
Chapter 5
In this area Meier et al . (1987) recognized a complex block mosaic with: (1) NE-trending dextral strike-slip faults; (2) SE-trending sinistral strike-slip faults; and (3) N-S compressive thrust faults. In the southern part of the study area right-lateral strike-slip motion of the Altamira Fault and the southern part of the EC is supported by the Cobbold et al. (1988) and Casero et al . (1995, 1997) interpretations that the southern EC is a zone of thrusting with minor right-lateral wrenching, mainly upon synthetic faults and right stepping folds and thrusts. In the Upper Magdalena Valley and southeastern part of the Cordillera transpressional displacement toward the NE with respect to the stable LLA area, suggested by the map view restoration, is supported by outcrop and seismic data f rom Bran Branquet (1999) and a nd Branquet Branquet et al. (1999b), (1999b), as well as by geophysical data by P ennington (1981, ( 1981, Fig. Daniels (1991, (1991, Fig. 5.18) and and Mora (1995, in Acosta et al .,., 2000, 2000, Figs. 5.2). 6.6 Chapter 6), Daniels Using field field and seismic data Branquet (1999) and Branquet et al. (1999b) (1999b ) interpreted interpreted a sec tion of the eastern flank of the Cordillera. They identified en-échelon folds in the h anging-wall anging-wall of major basement-involve asement-involvedd faults and they interpret basement pop-up structures limited to the east by convexupward reverse faults as positive flower structures. They proposed that these structures resulted from dextral transpressional inversion of an Early Cretaceous half-graben during the Andean orogeny. They also mentioned shallow-focus shallow-focu s earthquakes, shifting rivers and the morphology of the Llanos foothills (Fig. 2.1 for location) as e vidence of present-day dextral transpression. They also interpreted the Quetame Massif (Fig. 2.1) as as an en-échelon basement pop-up structure uplifted during tansspressional Andean deformation. This interpretatio int erpretationn is also supp orted by earthquake earthqu ake focal mechanism mecha nism solutions compiled by Pennington (1981, (198 1, Fig. 6.6 Chapter 6) and Daniels (1991, (199 1, Fig. 5.18). Global Global Positioning System data suggest a NE displaceme d isplacement nt of the Eastern Eastern Andes Bloc k relative to the stable South American Plate (Llanos; Mora, 1995, in Acosta et al .,., 2000, 2000, Fig. 5.2). Shortening and relative left-lateral displacement of blocks in the western flank of the EC (Magdalena-Tablazo su su b-basin) r ecognized ecognized in the map view restoration are supported by kinematic analysis of lineation data dat a on the main main f aults on t he western EC foothills and the Middle MV carried out by Acosta et al. (2000, Figs, 5.12 a nd 5.13). These authors suggested that: (1) the Bituima-La Salina and Chucurí faults ( Fig. 5.3) ar e represented by a series of steep ste ep dipping left-lateral le ft-lateral to oblique faults with a reverse sense; (2) The Cambras and Dos Hermanos faults fault s ( Fig. 5.3) ar e oblique faults with leftlateral and reverse sense that form a duplex. Cobbold et al . (1988) (1988) also suggested suggested that the northern EC is a zone of convergent left-lateral wrenching. The existence of two conjugate (?) or pseudo-conjugate (?) sets of NNW-SSE to NW-SE leftlateral strike-slip faults and NE-SW to E-W right-lateral strike-slip faults, is also confirme d from kinematic kinematic indicator fault striae measurements and other outcrop structural observations in the EC ( Figs. Cardozo-Puentes, 1989; Cardozo and Zúñiga, 1995; Kammer, 5.12 t o 5.15, Cobbold et al., 1988; Cardozo-Puentes, 1999b) and the the Táchira area ( Fig. 5.3; Meier et al., 1987). In the Santander Massif Mass if (along the th e Bucaramanga-Cúcuta road) Cobbold, et al. (1988) determined the Bucaramanga Fault (Fig. 5.3) and and other prominent faults with the same trend (340°) as reactivated Precambrian structures . These faults faults have oblique striae indicating dominantly left-lateral wrenching as well as minor thrusting. Another fault family striking 020° to 060° shows right-lateral thrusting. They hypothesised that the northern EC is a zone of convergent left-lateral wrenching. According to these authors the southern EC is a zone of thrusting with minor right-lateral wrenching, mainly upon synthetic faults and right stepping folds and thrusts. They speculated that the entire EC is at the restraining intersection of right- and left-lateral wrenches. 6.3.3. 6.3.3. Clockwise r otation of t he Centr Centr al Cor dil lera, M agdalena agdalena Val ley and wes wester ter n f lan k of th e Easte Eastern rn Cordill era
The clock-wise rotation of the Central Colombia Block (Central Cordillera), Magdalena Valley and western flank of the EC: Magdalena Tablazo sub-basin and western part of Cundinamarca sub basin relative to the stable South American Plate inferred from the map view restoration is in agreement with Acosta et al . (2000) interpretation. Acosta et al. (2000) using kinematic analysis of lineation data from the main faults on the western EC foothills and the Middle MV as well as regional interpretation of structures in the northern Central Cordillera and MV interpreted a left-lateral shear
211
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
0
500km
South Caribbean deformed Belt
CARIBBEAN PLATE
PANAMA BLOCK
Venezuela
A R E L L I D R O C R E T S A E
CHOCO BLOCK
I A B M O L NAZCA O C PLATE N R E T S E W S N I A R R E T I C N E A C Ecuador O
Colombia
SOUTH AMERICAN PLATE
Brazil
STRESS FIELD (P-AXES) INFERRED FROM FOCAL MECHANISMS IN THE UPPER 50 KM OF THE CRUST Straight lines indicate the trend of compression axes and length inversely proportional to the dip of the axis (from Daniels, 1991) Peru
ESTRUCTURES Thrust fault Strike-slip fault Normal fault
Figure 5.18. P-axes from earthquake focal mechanisms in the upper 50 km of the crust. Lines indicate the trend of the axis and the length is inversely proportional to the dip of the axis away from horizontal (from Daniels, 1991).
zone between the Bucaraman ga and Uramita faults formed as a result of indentation of the Chocó Block ( Figs. 5.1 and 5.2), resulting in transpression and clockwise block rotation (Acosta et al.,2 000). 6.3.4. Deformati on of the EC by tr anspressive basin in ver sion
Results of the map view restoration suggest that the SE flank of the EC (Cocuy and eastern Cundinamarca sub-basin) was right-lateral transpressively deformed but the NW flank of the EC
212
Chapter 5
(Magdalena-Tablazo sub-bassin) was left-lateral transpressively deformed (see for example thick broken black line Colletta´s section in Fig. 5.10). These results are in agreement with De Freitas et al. (1997) who interpreted that the EC co mprises: (1) A northern segment produced by the inversion of the Tablazo and Cocuy grabens separated by the Santa nder-Floresta palaeo-high. They recognized the Boyacá and Soapaga faults as inverted normal faults ( Fig. 2.28 Chapter 2); (2) A southern segment (Cundinamarca sub-basin) where the junction of the t wo northern depocenters implies accommodation of en échelon normal faults separated by strike-slip faults. They suggest that during basin inversion the western part of the Cundinamarca sub-basin was affected by N-S sinistral transpression, while the eastern part was affected by E-W dextral transpression. 6.3.5. Mi nor str uctur es r ecogni zed in outcr ops, but not i n th e map view r estorati on
Three structural features recognized from outcrop studies but not recognizable from the map view restoration are ( Figs. 5.13, 5.14 a nd 5.19): (1) Local NW-SE to N-S extension associated with minor normal faults crossing obl iquely to perpend icularly the structural grain of the EC. These local minor features have been recognized by Cardozo and Zúñiga (1995) and Kammer (1996, 1999b) in the eastern border of the Floresta Massif and by Cardozo-Puentes (1989) in the western part of the Cundinamarca sub-basin; (2) E-W to WNW-ESE-trending minor left-lateral strike-slip faults cross cutting the structural grain of the EC and thus of later (recent) origin. These faults have been recognized by Kammer (1999b), by Acosta et al . (2000) in the western flank of the EC and Middle MV, and by Cardozo-Puentes (1989) in the Sabana de Bogotá region; (3) E-W -trending right-lateral minor strike-slip faults cross cutting the major structures recognized by Cardozo and Zúñiga (1995) in the eastern flank of the Floresta Massif. Probably these structures are small enough to be detected in the regional map view restoration.
6.4. COMPARISON OF DISPLACEMENT OF BLOCKS RELATIVE TO STABLE SOUTH AMERICA WITH PLATE VELOCITY VECTORS FROM GEOPHYSICAL DATA Displacement of blocks relative to the stable cratonic South American Plate obtained from the map view restoration (Fig. 5.11) can be directly compared with velocity vectors determined from Global Positioning System measurements relative to South American plate (Kellogg et al .,1985; Freymueller et al ., 1993; Mora, 1995, in Acosta et al ., 2000; Fig. 5.2). Northeastward displacement vector s of the Garzón Massif relative to the stable South American Plate suggested by the map view restoratio n ( Fig. 5.11) are in notable agreement with velocity vectors obtained from GPS measurements (Kellogg et al ., 1985; Freymueller et al ., 1993; Mora, 1995, in Acosta et al., 2000, Fig. 5.2). Northeastward displacement vectors of the Upper Magdalena Valley-southeastern flank of the EC (eastern Cundinamarca sub-basin and Cocuy sub-basin) and southern Central Cordillera (south of the Ibagué Fault, Fig. 5.11) ar e in agreement with velocity vectors relative to stable South America from GPS measurements (Kellogg et al ., 1985; Freymueller et al ., 1993; Mora, 1995, in Acosta et al., 2000, Fig. 5.2). However, the gradual change toward E and decrease i n magnitude in the Cocuy area (Fig. 5.11) are in disagreement with the velocity vectors from GPS dat a ( Fig. 5.2). The WNW-ESE to E-W orientation of the regional stress field inferred from recent joints and stylolites (Mojica and Scheidegger, 1981; Mojica and Scheidegger, 1984; Mojica, 1985; Mojica, 1992, in Torres, 1992; Fig . 5.16), borehole breakout studies (Castillo and Mojica, 1990; Ochóa and Ponguta, 1991 and Torres, 1992, Fig. 5.17) and focal mechanisms solutions of earthquakes in the Andes Block (Fig. 5.18) support an internal deformation of the EC. Such an internal deformation of the EC is in a greement with the gradual change toward the east in the displacement vectors obtained from the map view restoration (Fig. 5.11). In contrast, earthquake focal mechanism solutions for the easternmost Andes b oundary faulted margin indicate an important dextral strike-slip component (Pennington, 1981 , Fig. 6.6 Chapter 6) which is in agreement with the Northeastward velocity vectors obtaine d from GPS data (Ke llogg et al ., 1985; Freymueller et al ., 1993; Mora, 1995, in Acosta et al., 2000, Fig. 5.2). In general for the
213
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
0
Serrania de Perija
100 km
Maracaibo Basin
B u c a r a m a n g a F
s a c u L n a S e d ia n a e r r S
r a e l l i r d o C
u e F I b a g
Barinas basin
Arauca
F n da o H
Manizales
Cucuta
Bucaramanga
F F a s i n l n o a S a m a r L e H s o D
Medellin
a r i d e M
s d e n A
F r e z u a S
Llanos Orientales Basin (LLA)
Tunja
F a m i t u B i
A E R L L D I R C O N R T E S E A
Bogota
Yopal
Ibague
Villavicencio
l r a t e n C
Guyana Shield
Neiva
y le V a a f i s a le n a s g d M a o n z e r M a r pp G U
a e n d e r a a i n c a a r M r e a S L
OUTCROP STRUCTURAL DATA 2. Acosta et al. (2000)
Figure 5.19. Kinematic indicators of northwest-trending left-lateral strike-slip faults (from Acosta et al., 2000).
214
Chapter 5
study area, the conflict between the displacement vectors from the map restoration (Fig. 5.11) and the velocity vectors from GPS data (Figs. 5.2), can be resolved if we consider that the di splacement vectors from the map view restoration represent mostly the Neogene Andean deformation and the GPS velocity vectors represent the present-day movement of the Andes relative to stable South American Plate. Probably the Andes Block as a whole is at present moving with a dominant right-lateral strike-slip component along the faults in its easternmost boundary. This strike-slip component was probably much less important during the whole period of Neogene Andean deformation as suggested by the map restoration.
6.5.
ANDEAN NEOGENE DEFORMATION, BASIN INVERSION OF THE EASTERN CORDILLERA AND LITHOSPHERE RHEOLOGY
Andean deformation of the EC and its northern prolongations, th e Sierra de Perijá a nd the Mérida Andes, coincide with the location of Mesozoic extensional basins ( Figs. 2.13, 2.15 a nd 2.17; Chapter 2). These extensional basins were more intensely deformed as mountain belt s than the more rigid blocks with Neogene sedimentary basins surrounding them ( Figs. 1.1, 5.5 an d 5.7). Young rifts are characterized by weak lithosphere ( e.g. Ebinger et al., 1989, in Burov and Diament, 1995) due to thermal weakening and necking of the lithosphere. Weak thermal destabilized lithosphere in the area of the former extensional basins (Jurassic to Early Cretaceous rifts) was prone to deformation. Transpressive tectonic stresses generated inversion of these former extensional basins to generate these mountain ranges. Under transpressive tecto nic stresses the major normal fault systems originally delimiting the extensional basins were invert ed ( Fig. 5.4). In the external western and eastern foothills of the EC short-cuts within the sedimentary c over generated thrust systems (Fig. 5.4).
7. CONCLUSIONS The Northern Andes of Colombia and Venezuela are located in a broad zone of deformation resulting from the interaction of the South American, Nazca and Caribbean plates. Such interaction has fragmented the NW corner of the South American Plate into a number of micro-plates or tectonic blocks: Guajira, Santa Marta, Maracaibo and Central Colombia (Colombian Central Cordillera). Relative movements of the tectonic blocks resulted in deformed belts with a combination of compressional thick/thin-skinned thrusting, folding and strike-slip faulting which form several mountain ranges or strike-slip fault zones in the Northern Andes. I have manually restored in map view a mosaic of fault-bounded blocks using all the kinematic constraints available in literature, such as strike-slip displacement estimates and amounts of shortening from available balanced cross-sections. In order to simplify the model I have used a limited number of blocks. This technique is useful to constrain shortening and strike-slip displacement estimates as well as to detect strike-slip motions and rotations about vertical axes not revealed by balancing cross-sections. According to this regional map restoration, the following relative movements of regional blocks leading to deformation belts were interpreted: (1) Eastward convergence and clockwise rotation of the Central Colombia tectonic block relative to South America resulted in transpression of the EC, with northward increasing shortening in this mountain range; (2) Eastward convergence and left-lateral strike-slip movement of Central Colombia relative to Maracaibo produced the northern part of the EC in the Santander Massif region; (3) Northeastward convergence of the Santa Marta Block relative to Maracaibo generated the Perijá Mountain Range; and (4) Southeastward convergence and right-lateral strike-slip motion of the Maracaibo Block relative to South America generated the Mérida Andes of Venezuela. A more detailed map view restoration of the EC of Colombia was done using balanced crosssections from the literature. According to the restoration the amount of shortening during Andean deformation is approximately one half of the present-day width of the Cordillera. Both width and amount of shortening increase northward. Results of the map view restoration are in general supported by kinematic indicators (fault striae) and other outcrop structural data, as well as by the orientation of stress field inferred from recent vertical joints, stylolites, borehole breakout data and the focal mechanism solutions of upper crustal earthquakes. These results indicate: (1) ENE-WSW shortening
215
Map View Restauration of Transpressional Basin Inver sion in the Eastern Cordillera
perpendicular to the regional structural grain of the EC; (2) Conjugate (?) or pseudo.conjugate (?) leftlateral and right-lateral strike-slip faults; (3) Clockwise rotation of the Central Cordillera, Magdalena Valley and western flank of the EC; (4) The SE flank of the EC (Cocuy and eastern Cundinamarca sub basin) was right-lateral transpressively deformed; (5) The NW flank of the EC (Magdalena-Tablazo sub-basin) was left-lateral transpressively deformed; (6) Andean deformation generated the EC through transpressive inversion of Mesozoic extensional basins. The western part of the EC was affected by NS sinistral transpression, while the eastern part was affected by E-W dextral transpression (see for example thick broken black line Colletta´s section i n Fig. 5.10). Comparison of displacement of blocks relative to stable South America obtained from the map view restoration, with plate velocity vectors from geophysical data, suggest that the Andes Block as a whole is at present time moving with a dominant right-lateral strike-slip component along the faults at its easternmost boundary. This strike-slip component probably was less important during the whole period of Neogene Andean deformation, as suggested by the map view restoration.
216
CHAPTER 6 RHEOLOGICAL EVOLUTION OF THE LITHOSPHERE OF THE EASTERN CORDILLERA AND HYPOTHESES ABOUT ITS DEEP STRUCTURE 1. INTRODUCTION Two major controls on lithosphere dynamics of basin formation and inversion are: (1) lithosphere rheology and (2) stresses affecting the lithosphere (Ziegler et al ., 1995, 1998). The aim of this chapter is to constrain the Meso-Cenozoic lithospheric scale tectonic evolution of the EC in terms of the interaction of rheology of the lithosphere and stresses affecting it. Application of this concept helps to understand the present-day structure of the EC in terms of its Meso-Cenozoic tectonic evolution, and allows one to propose some hypotheses about its deep structure. In the first part of the chapter I summarise the available geophysical data used to constrain the deep structure of the EC. Then I review rheological properties of the continental lithosphere and present rheological models of the EC lithosphere along a regional cross-section through its Mesozoic extensional basin formation and Cenozoic basin inversion history. Using the results of these rheological models and previous flexural models I discuss the rheological evolution of the EC lithosphere during Meso-Cenozoic time. Subsequently I discuss the probable plate-tectonic-related palaeo-stress field evolution that affected the EC lithosphere and the interaction with the EC lithosphere throughout the Meso-Cenozoic tectonic evolution of the EC. Finally, I compare the EC with similar mountain belts, and with analogue and numerical model experiments from literature in order to propose some hypotheses about the deep structure of the EC.
2. GEOPHYSICAL DATA CONSTRAINING THE DEEP STRUCTURE OF THE EASTERN CORDILLERA Data constraining the present-day deep structure of the EC are limited to seismic refraction, gravity and earthquake data.
2.1. SEISMIC VELOCITY MODEL BASED ON REFRACTION IN THE SOUTHWEST OF COLOMBIA Ocola et al . (1975) presented a model of seismic velocity in the crust and u pper mantle, f or a refraction line going from La Cocha Lake in the southern part of Colombia to Bogot a (Fig. 6.1). Leeds (1977) used this seismic refraction model and with additional surface wave meas urements using a resolution analysis (Wiggins, 1972) calculated a shear wave velocity model for the crust and mantle in the Colombia-Ecuador region (Fig. 6.1). According to this model there is a zone of low seismic velocity at a depth of 110-200 km. For simplicity Coral (1985) assumed that the low velocity zone extends below all of the EC.
2.2. GRAVITY AND MOHO DISCONTINUITY DEPTH I used gravity data from the map of Colombia, Eastern Panama and adjacent marine areas published by Kellogg et al . (1991). The depth to the Moho in Colombi a (Fig. 6.2) has been calculated by Salvador (1991) using the following method: (1) Application of free air, Bouguer and topographic corrections to the gravity data; (2) Two-dimensional gravity modelling along three seismic refraction profiles, in the west and southwest part of Colombia (Ramírez and Aldrich, 1973; Ocola et al ., 1975; Money et al ., 1979; Flüeh et al ., 1981), in order to isolate the gravity effect of the Moho. He found the frequency effect of the Moho fell in the 250-580 km wavelength band; (3) The gravity field due to the Moho was recovered from the total field using a two-dimensional band pass filter with these corner frequencies. The estimated depth to the Moho was consistent with the values inferred from refraction seismic. Also Salvador (1991) applied flexural models considering only the topographic load of the EC and assuming
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
A.
Velocity (km/sec)
3
4
5
6
7
8
9
50
)
mk ( th
100
p e
D Vp 150
200
Vs
B.
5o E C O
4o
A
N
BOGOTA BUENAVENTURA
3o
2o
1o
P
A
C
I
F
I
C
COLOMBIA
PASTO LA COCHA
0o QUITO P
1o
E R U
E C U A D O R RIO BAMBA
2o
GUAYAQUIL
81o
80o
79o
78 o
77o
76o
75o
74o
73o
Figure 6.1: A. Seismic velocity model of the crust and upper mantle along a refraction line going from La Cocha Lake in the southern part of Colombia to Bogota in the EC. Vp: compressional wave velocity (from Ocola et al , 1975); Vs shear wave velocity, the arrows represent the uncertainties for the two mantle layers (from Leeds, 1977). B. Location of the refraction line (from Ocola et al , 1975).
218
Chapter 6 3 3
3 3
s a u c L n a S e d ia n r ra e S
3 4
1400 3 5
36
1300
37
3 5 3 7 3 9
3 6 3 8
41
4 2
38
4 4
4 0
4 5
4 0 4 1
4 4 4 2
Medellin
3 3
Arauca
4 5 44 4 3 4 2
4 3 4 2
4 3
4 4
ra ill e rd o C
1100
basin
7 6 3 9 3 9 3 8 3 3 5 3 3 4
4 1
4 5 4 4
3 9
Barinas
46
Bucaramanga
38
1200
s d e n a A r i d e M
Cucuta
4 0
4 3
4 1 4 0
Maracaibo Basin
Serrania de Perija
3 4
Manizales
r a l e l i 44 d o r C 45 r n Tunja e 46 s t a E
4 1
47
Yopal 4 0
48
3 9
Llanos Orientales Basin (LLA)
3 8
1000 3
900
3
4 3 5 3
3 6
Ibague
37
8 3
43 44 45 4 6 4 7 4 4 6 4 8 7
800
4 4
3 9
l a t r n e C 4 1 4 2
800
7 3
Bogota 4 7 4 6
3 6 Villavicencio
4 5
4 3
4 0
4 2
3 0 2 9 2 8
3 7 5 4
4 4
3 4
100
200 Km
32
4 2
Neiva
3 4
0
33
4 1
5 3
4 2
900
4 1
4 0 3 9 3 8
S
1000
3 6 5 4 3 3 3 3
MOHO DEPTH
2 7
San Jose del Guaviare
(km) from gravity
3 1 2 1 3 3
1100
1200
1300
1400
Figure 6.2: Moho discontinuity depth under the study area estimated from gravity data (modified after Salvador, 1991). Major strike-slip faults and suggested subcrustal mantle lithosphere subduction under the EC are also shown (see discussion in section 6.3, 6.4a nd 6.5).
a laterally constant effective elastic thickness. He obtained values of 25 km which is in excellent agreement with the values for most of the EC presented here, which were calculated assuming lateral changes in elastic thickness. Salvador (1991) interpreted the presence of a volcanic chain in the Central Cordillera as the cause for a weak lithosphere, and he also found that the gravity effect of the deflection is enough to compensate the topography.
219
1500
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
2.3. SEISMICITY Figure 6.3 shows seismicity distribution of the EC, obtained from the United States Geological Survey, National Earthquake Information Center. Three profiles of seismicity across the Colombian Andes are shown in figure 6.4 (f rom Pennington, 1981) and two profiles of seismicity across the EC are shown in figure 6.5 (from Taboada et al ., 1999; 2000).
Serrania de Perija
Maracaibo Basin Cucuta
8
B´ Barinas basin
Bucaramanga Arauca
7
A Medellin
6
Paipa Tunja Iza Yopal
A´
Manizales
5
B Bogota
Llanos Orientales Basin (LLA)
Ibague
Villavicencio
4
Guyana Shield
3
Neiva
0
100
200 Km
San Jose del Guaviare
-76
-75
-74
-73
-72
-71
-70
Figure 6.3: Seismicity in the study area. Triangles: shallow earthquakes (< 10 km). Lozenges: intermediate-depth earthquakes (10 to 60 km). Circles: Deep earthquakes (>60 km). (modified from United States Geological Survey, National Earthquake Information Center. World Data Center for Seismology, http://wwwneic.cr.usgs.gov/neis/epic/epic.html). Note the local presence of volcanic rocks at Paipa and Iza and the Bucaramanga earthquake nest south of Bucaramanga. AA´ and BB´ represent the location of seismicity crosssections shown in figure 6.5.
220
Chapter 6
SE
NW 20
100
D 200
C a ( P e r i b b e r e z a n P e t a l . , l a t e 1 9 9 7 )
a BUCARAMANGA SEGMENT
100 KM
300
NW
SE
SW
NE
35
( T N a b a o a z c d a a P e t l a t a l e b . , 2 0 0 CAUCA SEGMENT 0 )
35
100
D 200
c ECUADOR SEGMENT
300
Figure 6.4: Cross-sections of the seismicity of the three segments of subducted oceanic lithosphere, including the seismicity of the overlying continental lithosphere. No vertical or horizontal exaggeration. Larger circles are excellent and high quality locations. The positions of volcanoes projected onto the line are shown at the top of each diagram, where they exist. The Bucaramanga nest is shown as a solid symbol in the top diagram. Location of the sections is in Figure 6.8 (f rom Pennington, 1981 and after Pérez et al ., 1997 and Taboada et al ., 2000). Note the subduction angle (20°) of the Caribbean Plate north of Colombia in (a) and the steeper angle of the Nazca Plate (35°) in (b).
On the base of well located seismic events ( Fig. 6.4) and focal mechanism solutions ( Fig. 6.6) Penington (1981) concluded ( Fig. 6.7) that: (1) The Panama block is accommodating east-west compression along a series of thrust faults striking NW to N E; (2) The Andean Block (Ecuador, Colombia and Venezuela Andes) is moving as a block NNE relat ive to South America, as indicated by right-lateral strike-slip and thrusting focal mechanism solutions ( Fig. 6.6), along a system of faults following the eastern front of the EC; (3) The subducted portions of the old Farallon Plate under the Colombian Andes have been segmented into three parts ( Fig. 6.7): (a) A Bucaramanga segment continuous with the Caribbean plate NW of Colombia; (b) A Cauca segment continuous with the Nazca Plate being subducted west of Colombia-Ecuador; and (c) An Ecuador segment at the northern end of the subducted lithospheric plate which is dipping at a small angle to the east beneath northern Peru. However, according to Isacks and Molnar (1971) subduction is dipping to the east beneath Ecuador and part of Colombia and to the south-east beneath northern Colombia. Pennington (1981) also recognised that the Bucaramanga and Cauca segments can alternatively be interpreted as one continuous contorted segment, of shallower dip to the north and with some aseismic regions within it. A particular feature of the seismicity of the EC is a small zone ( <65 km in diameter) of intense seismicity centred at a depth of 160 km, known as the Bucaramanga earthquake nest (Schneider et al ., 1987).
221
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Figure 6.5: Seismicity section across the EC. Location of the section in Fig.6.3. No vertical or horizontal exaggeration. Vertical exaggeration for topography is 3, maximum topogra phic elevation is 4 km. (a) Section orthogonal to the EC. It is appr oximately the sa me of the structural balanced cross-section of the EC by Colletta et al . (1990; Figs. 5.4 and 5.7 Chapter 5, and Figs. 6.13 and 6.27). (b) Section parallel to the axis of the EC. Earthquakes have been projected from a distance up to 150 km from the section (a) and up to 100 km from the section (b). Note the seismicity below the EC suggests a subduction angle about 35°-45° greater than the Caribbean subduction angle below northern Colombia (see Fig. 6.4 and 6.8). (f rom Taboada et al ., 2000).
222
Chapter 6
15oN
Caribbean Plate
FFP 10o
Bloc andin
B F
Cocos Plate
South American Plate U F G
Nazca Plate
h c n e r t n a i b m o l o C
R G F D A F G
0o
FDGR : FGA : FGU : FB : FEP : 85oW
80o
5o
75o
Romeral Fault Garzón-Altamira Fault Guaicaramo Fault Bocono Fault El Pilar Fault 70o
5oS
65o
Figure 6.6: Focal mechanism solutions along the eastern border of the Andes (Andean Block; after Pennington, 1981, Súarez et al ., 1983, Cisternas et al ., 1984 and Guillande, 1988).
Based on new data from the National Seismological Network of Colombia (NSNC), during the period 1993 to 1996, Taboada et al . (1999, 2000) concluded that seismicity of the EC has the following features: (1) Crustal earthquakes concentrated along the eastern and western foothills of the Cordilleras are aligned with the main fault s (Fig. 6.5). Shallow seismicity ( <2 5 km), occurs in the brittle domain of the upper crust. Taboada et al . (1999) interpreted the scarce seismicity between 25 and 55 km depth as coming from the lower crust ( Fig. 6.5).; (2) Intermediate seismicity shows a NNE trendi ng subduction segment beneath the EC with a well-defined cluster along the NW margin of the Cordill era. (Fig. 6.5) This subduction segment was not well defined before the installation of the NSNC. Taboada et al . (2000) suggested that this slab corresponds to a remnant of the Palaeo-Caribbean Plate attached to the
223
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Baudó-Panamá Block which displays a shallow subduction beneath the EC (Southern part of the Bucaramanga segment of Pennington, 1981, Fig. 6.7). 15 o N
Caribbean Plate
5 0 0 1 0 0 1 0
Cocos Plate
0 0 0 1 0 8 4 1 1
Nazca Plate
0 5 1
Caribbean Plate northern Bucaramanga segment
BN
a
0 0 +Palaeo-Caribbean Plate 0 2 southern Bucaramanga 1 4 0 1 8 segment + 0 0 2 Nazca Plate b Cauca segment
Ecuador 2 0 0 + 1 8 1 4 0 1 0 0 0 80 o
5o
0o
c
o 85 W
10 o
75 o
segment
70 o
4 oS 65 o
Figure 6.7: Contours to the Benioff zones beneath north-western S outh America (labelled in kilometres). Lines a, b and c show the location of seismicity cross-sections shown in figure 6.4. These three segments of subducted lithosphere: Bucaramanga segment, Cauca segment and Ecuador were identif ied by Pennington (1981). A fourth fragment of the Palaeo-Caribbean Plate attached to Panamá under the EC has been recently demonstrated by Taboada et al ., (2000, see Fig. 6.5). Note that the strike (N 30° E) and subduction angle ( 35° to 45° SEE) of the Palaeo-Caribbean Plate fr agment beneath the EC (from Pennington, 1981 and Taboada et al ., 2000) are different from the strike (N 40-45° E) and subduction angle (20° SE) of the Caribbean Plate beneath northern Colombia (from Pennington, 1981, and Pérez et al ., 1997). The different geometry of these slabs fragments and the different velocity vectors suggest that they are converging obliquely along a SEE boundary where the Bucaramanga earthquake nest (BN) is located (see also Fig. 6.9)
The intermediate seismicity slab beneath the EC is limited southwards, along a roughly E-W trending boundary approximately located at 5.2 N ( Fig. 6.7). This boundary aligns with the southern termination of the Baudó Range, and Taboada et al . (1999, 2000) suggested that this shear zone defined
224
Chapter 6
by the Itsmina and Ibagué Faults and the Santamarta-Bucaramanga Fault are the b oundaries of a moving continental wedge associated with the most active segment of this subduction ( Figs. 6.7 an d 6.8). The formation of this subducted slab probably is linked with the accretion of the C hocó Block, the indentation of the continental wedge between the Itamina-Ibagué Faults and the SantamartaBucaramanga Fault and the uplift of the EC (Taboada et al ., 2000). Accretion of the Chocó Block at 12
20 Ma. Palaeo-Caribbean Plate
Palaeo-Cocos Plate o a
Pa laeo -Nazca Pl ate
a
a´
Present
80 W
Carib be an Pl ate 10
b
Cocos Plate
b´
5
b
5 0 0 km Naz ca P la te
Figure 6.8: Tectonic plate reconstruction of the Northern Andes. (a) Reconstruction at 20 Ma (modified from Taboada et al ., 2000). (b) Reconstruction at present time (modified from Taboada et al ., 2000). The collision of the Baudó-Panamá island arc (BPA oblique line pattern) favouring the uplift of the EC began at 12 Ma. Black circles indicate the location of active volcanism. Shaded areas indicate mountain ranges. At 20 Ma the Central Cordillera (CC) and low topography mountains produced by incipient inversion of Mesozoic extensional basins (TS Tablazo inverted sub-basin, CS Cocuy inverted sub-basin, SM Santander Massif, MA Mérida Andes, GM Garzón-Quetame Massif were uplifted areas. EC Eastern Cordillera, WC western Cordillera, UMV Upper Magdalena Valley, MMV Middle Magdalena Valley, LMV Lower Magdalena Valley, SSJ Sinú-San Jacinto fault and fold belts, AR abandoned ridge.
225
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
0
Figure 6.8: Continued (c) Schematic tectonic cross-sections of the Northern Andes at 20 (section aa´) and at present time (section bb´). NP Nazca Plate, PCP Palaeo-Caribbean Plate, RFS Romeral Fault System, CML, continental lithospheric mantle, WC Western Cordillera, CC Central Cordillera, MV Magdalena Valley, EC Eastern Cordillera (from Taboada et al ., 2000). 80° W 10° N Caribbean Plate
Panama
SJ J
5° N Coiba Micro-plate Nazca Plate WC
SNSM SP
LM V CC
70° W
MA
MM V
EC L LA
South Amer i can Plate
BN k 1
Figure 6.9: Schematic block diagram indicating the 3D geometry and kinematics of the lithospheric plates beneath the EC and north-western Colombia. Intracontinental subductions are defined beneath the EC and the Mérida Andes. The Bucaramanga nest (BN) is located at the oblique collision between the Palaeo-Caribbean subduction segment beneath the EC connected to the Panamá Block and the Caribbean Plate subducting beneath northern Colombia. Very deep overturning of the Nazca Plate has bee interpreted by Taboada et al . (2000) from mantle tomographic sections. the inflected zone where the EC subduction joints the Santamarta-Bucaramanga Fault (from Taboada et al ., 1999). EC Eastern Cordillera, LLA Llanos Basin, MMV Middle Magdalena Valley, LMV Lower Magdalena Valley, CC Central Cordillera, WC Western Cordillera, MA Mérida Andes, SP Sierra de Perijá, SNSM Sierra Nevada de Santa Marta (modified after Pennington, 1981; Pérez et al., 1997; Taboada et al ., 1999, 2000).
226
Chapter6
Ma blocked normal oceanic subduction of the Caribbean Plate beneath NW South America ( Fig. 6.8; Duque-Caro, 1990; Taboada et al ., 2000). Taboada et al ., (2000) interpr eted that the northern part of the Nazca Plate (Cauca segment according to Pennington, 198 1; Fig. 6.7) subducts beneath the Chocó Block overlapping the southern part of the Palaeo-Caribbean Pl ate fragment recognized beneath the EC (southern part of Bucaramanga segment of Pennington, 1981; Figs. 6.7 t o 6.9). The convergence rate along the oceanic trench decreased and active deformation shifted towar d weak zones of conti nental lithosphere. Shortening localised along Mesozoic extensional basins, crea ting tectonic inversion ( Fig. 6.9). New data from the NSCN suggests that the Bucaramanga earthquake nest is not an isolated phenomenon as previously though. Taboada et al ., (2000) suggested that it is related to an inflect ion of the subducting surface. However the subduction angle (~35 – 45°) of the Palaeo-Caribbean platefragment below the EC is stepper than the Caribbean plate subdu ction beneath northern Colombia (about 20° according to Pennington, 1981 and Pérez et al ., 1997). (see Figs. 6.4, 6.5, and 6.7).
2.4. GEOTHERMAL REGIME A compilation of geothermal gradient and heat flow estimates from literature is presented in figures 6.10 an d 6.11. Puerto (1982), Bachu et al . (1995) and ICP-GEX-DYA (1995) studied the geothermal regime of the Magdalena Valley, Llanos Orientales and Neiva sub-basin (Upper Magdalena Valley) respectively. According to these authors these data should be considered preliminary because bottom hole temperatures are incomplete and contain errors, and thermal conductivity has not been measured. The few estimates for the EC should also be considered preliminary. These estimates have been calculated from organic matter maturity models using bottom hole temperatures and vitrinite reflectance data by Mora (1996, 2000) and Lance-Le Cornec (1998). Local lateral changes in geothermal gradients and heat flow estimates may result from convective heat transport due to water circulation within the sedimentary rocks (Bachu et al ., 1995; ICP-GEX-DYA, 1995) and evidenced by numerous hot springs in the EC (Forero, 1958). Hydrothermal activity (300°C) has also been interpreted during emerald formation in the emerald mines (western and eastern flanks of the EC) by Cheilletz et al ., 1993). Such hydrothermal activity related to emerald crystallization occurred during Palaeogene (38-32 Ma) according to Cheilletz et al . (1993) based on 40Ar/39Ar and K/Ar data, or close to the Cretaceous-Paleocene boundary (67-61 Ma) according to Romero et al . (2000) based on 87Sr/86Sr data.
3. RHEOLOGICAL EVOLUTION OF THE LITHOSPHERE OF THE EASTERN CORDILLERA 3.1. RHEOLOGY OF THE CONTINENTAL LITHOSPHERE The assessment of lithospheric rheology is an important tool in understanding lithospheric dynamics (Ranalli, 1995). In this section I review the rheology of the continental lithosphere and I present rheological models of the lithosphere of the EC along a regional cross-section through Mesozoic and Cenozoic time. 3.1.1. Rheology of th e conti nental l ith osphere fr om r ock mechani cs data
The following summary about the rheology of the lithosphere has been extracted from Beekman (1994) and Van Wees (1994). The rheological behaviour of rocks that constitute the
227
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
continental lithosphere is derived from laboratory experiments and subsequently extrapolated to geologically relevant times. At low confining pressures and low temperatures, such as those of the upper crust, brittle fracture is predominant (strain-rate independent deformation). Fracturing is described by a MohrCoulomb criterion, which, following Byerlee (1978), can be expressed in terms of effective principal stress difference, lithostatic overburden pressure, and pore fluid pressure (Ranalli, 1995):
1400
1300
Serrania de Perija
GEOTHERMAL GRADIENT (°F/100 feet)
Barinas
Cucuta
basin
After Puerto (1982) Bachu et al. (1995) ICP-GEX-DYA (1995) Lance-Le Cornec (1998)
1200
Maracaibo Basin
100
0
0.75
200 Km
Bucaramanga Arauca
Medellin
2.90 Tunja 3.12 3.34
1100
1.25
2.00
3.51 Yopal
3.51 Manizales 2.52 Bogota
1000 Ibague
3.62
900
2.25
1.75
3.95 1.25 1.25 Villavicencio 3.45 3.40 1.50 3.29 3.73 2.76 3.01
2.50
Neiva
2.75
3.00
800
Guyana Shield
San Jose del Guaviare
800
900
1000
1100
1200
1300
1400
15
Figure 6.10: Compilation of present day geothermal gradient estimates. Estimates for the Llanos Basin from Bachu et al . (1995), for the Middle and Upper Magdalena Valley from Puerto (1982), some estimates for the Upper Magdalena Valley from ICP-GEX-DYA (1995) and for the EC from well data and organic matter maturity models from Mora (1996, 2000) and Lance-Le Cornec (1998).
228
Chapter 6
1400
Serrania 35.0 Maracaibo de Perija Basin
THERMAL FLOW (mW/m 2 )
s d e n a A d i r M e
Afte r B a c h u e t a l. (1 9 9 5 ) I C P -G E X -D Y A (1 9 9 5 ) Lanc e- Le Corne c (199 8) 1300
Barinas
Cucuta
basin
0
Mora (1996, 2000)
100
200 Km
Bucaram anga Arau ca
50.0
r e l Medellin l i a d
1200
r
47.0 Tunj a 60.0 39.0
o 1100
C
50.0 Yopal
47.0 Manizales
30.0
35.4 29.2 26.8 37.2 35.4
51.9 38.1
28.0 Bogota
1000
38.8
33.0
Ibague
9 00
48.5
33.6 33.0
l a 68.3 t r n e 5 7 .6 57 . 9 C 53.9
38.6
41.4
58.5
62.7
Villavicencio
42.0 34.0 34.0
55.0
36.0 19.0
54.4
65.3 262.1 54.7 Neiva
8 00
313.7
Guyana Shield
San Jose del Guaviar e 62.3 35.0
8 00
9 00
1 00 0
1 10 0
1 20 0
1 30 0
1 40 0
1 50 0
Figure 6.11: Compilation of present day geothermal heat flow estimates. Estimates for the Llanos Basin from Bachu et al . (1995), for the Middle and Upper Magdalena Valley from Puerto (1982), some estimates for the Upper Magdalena Valley from ICP-GEX-DYA (1995) and for the EC from well data and organic matter maturity models from Mora (1996, 2000) and Lance-Le Cornec (1998).
σ brittle
= σ 1 − σ 3 = αρ gz (1 − λ )
where
R − 1 R α = R − 1 R − 1 1 + β ( R − 1)
229
For normal faulting For thrust faulting For strike-slip faulting
(6.1)
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
where σ1 and σ3 are maximum and minimal principal stresses, ρ is density, g is gravitation acceleration, z is depth, and where
λ
=
P 1
ρ gz
=
ρw ρ
≈ 0.35 is the hydrostatic pore fluid factor
0 < β < 1 denoting magnitude of intermediate stress σ2 = σ3 + β(σ1 - σ3)
R
= [ 1 + µ − µ ]−2
µ is the (static) sliding friction coefficient. According to Ranalli (1995) a common value is µ = 0.75
Steady-state creep of a wide variety of rocks, as believed to occur in the lower crust, is described by a ductile flow law. Experiments with rock-forming minerals show that the critical principal differential stress necessary to maintain a given steady-state strain rate, is a function of a power of the strain rate and varies strongly with mineral composition and temperature. The ductile flow obeys the so called power-law creep (Kirby, 1983): 1
σ creep
ε n E p = σ 1 − σ 3 = exp A p nRT
(6.2a)
Where ε& is strain rate A P , n and E P are flow parameters, T is absolute temperature and R is the gas constant. The above power-law creep indicates that critical differential principal stresses grow exponentially with decreasing temperature. Flow parameters for various rock-forming minerals show that power-law creep flow parameters are strongly controlled by silica content ( e.g. Carter and Tsenn, 1987). Felsic rocks ( e.g. granite) show low critical differential stress values compared to mafic rocks (e.g. dunites), under similar conditions of strain rate and temperature. The flow parameters are also very sensitive to the abundance of water. Wet rock samples show significantly lower critical differential stress values than dry ones. However, according to the experiments, above a critical stress, the power-law creep breaks down into the so-called low-temperature plasticity, which is characterised by a nearly linear increase in stress with decreasing temperature. For olivine, the main constituent of the subcrustal lithosphere, Goetze and Evans (1979) found that at differential stresses higher than 200 Mpa (=2 kbar) low-temperature plasticity is described by the so-called Dorn law:
σ creep
1 2 RT ε ln = σ 1 − σ 3 = σ D 1 − − E A D D
Where A D , E D, and σ D are flow parameters for low-temperature plasticity.
230
(6.2b)
Chapter 6
3.1.2. Rheological prof il es and i ntegrated str ength of a strati fi ed li thosphere
For a given tectonic environment (thrusting, normal faulting or strike-slip faulting), at any depth, flow properties, temperature and strain rate, the lowest of the brittle and creep principal stress difference given by equations (6.1) and (6.2a/b) respectively, gives a rheological strength (called the yield strength). For stress differences below this yield strength, the imposed strain rate will not occur. The dependence of the principal stress differences in equations (6.1) and (6.2a/b) on external conditions (pressure and temperature) and material properties (mineralogical composition), both varying with depth within the lithosphere, makes the yield strength also vary with depth, thus constituting a strength envelope. Continental crust, with a thickness in the order of several tens of kilometres, is layered, with large variations between different tectonic provinces. A simple and representative crustal mineralogical model is a two layered crust with felsic rocks dominating the upper crust and mafic rocks dominating the lower crust, as indicated by seismic velocities. These crustal layers overly ultramafic upper mantle material composed primarily of olivine. For a typical continental lithosphere composed of a quartzite upper crust, a diorite lower crust, and an olivine mantle, the associated rheological layering is shown in Figure 6.12. Composite strength envelopes are shown for compressive (thrust faulting) and tensile (normal faulting) stress regimes at several stages of thermal cooling. Lithosphere strength envelopes typical for Phanerozoic extension and inversion settings show a marked layering of a relatively strong (mostly brittle) upper crust, a weak (mostly ductile) lower crust, and a strong (brittle and ductile) subcrustal lithosphere. This layering, predicted from extrapolation of rock mechanics, agrees quite well with interpretation of geophysical and geological data. This rheological layering is confirmed by the depth distribution of seismic activity, which is approximately confined to the crustal layers ( e.g. Cloetingh and Banda, 1992), as well as by the presence of minima in seismic wave velocity and electrical resistivity coinciding with the ductile layers (Ranalli and Murphy, 1987). The rheological stratification may significantly affect the thermo-mechanical response of the continental lithosphere to stresses. For example, large normal faults in the upper crust tend to flatten out towards a detachment zone in the lower weak part of the upper crust (Braun and Beaumont, 1987). Another possible effect of the strong/weak intracrustal alternation may be mutual mechanical decoupling of the layers under substantial loading along intra-lithospheric shear zones leading to breakdown in coherent deformation of the multi-layered continental lithosphere (Ord and Hobbs, 1989). A scalar measure for the total strength of a (multi-layered) lithosphere with a depth varying rheology can be obtained by vertically integrating the yield envelope: h
σ L
= ∫ σ y ( z )dz
(6.3)
0
Where σ L represents the total lithospheric strength, h is the thickness of the mechanically strong lithosphere, defined, for example, by the depth where the rheological yield stress has decreased to a level of 10 Mpa and with no further rheological discontinuities at larger depths.
231
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure Strength (MPa) / Temperature 0 0 0 1 0
0 0 5 - 0
ext
0 0 5
0 0 0 1
Strength evolution of rifted lithosphere Sediments
comp
Crust
20
40 60 Km
Mantle Lithosphere 100 Ma
90 Ma
30 Ma
60 Ma
0 Ma
Coolng
Stretching
Age (Ma) ) m 25 / N T ( 20 h t g 15 n e r t s 10 d e 5 t a r g e 0 t n I
100
80
60
40
20
0
Compression
Extension
Figure 6.12: Depth-dependent rheological models for the evolution of a stretched lithosphere (from Van Wees, 2000).
3.1.3. Thermal stru ctur e of conti nental l ith osphere
Temperature plays an important role in the thermo-mechanical response of the lithosphere to stresses. Of particular importance is the variation of temperature with depth on the ductile strength of lithospheric rocks. Conductive cooling of the lithosphere over a relatively short period of a few million years after the last heating event rapidly leads to a significant increase in the mechanical strength of the lithosphere. Over longer periods cooling also leads to thickening of the lithosphere, and of its mechanically strong sub-layers. For both oceanic and continental lithosphere the derived geotherms are solutions of the Fourier’s law of heat conduction
ρc p
∂T = ∇.(k ∇T ) + A ∂t
(6.4)
where T is absolute temperature, t is time, and with the material parameters: ρ density, C p heat capacity at constant pressure, k the coefficient of heat conduction, and A the radioactive heat production per volume unit. Material parameters are assumed to stay constant within the major crustal and subcrustal layers, except for the heat production A, for which in some cases an exponential decay with depth is adopted. Initial geotherm (temperature-depth curve) were assumed to be equal to those of a steady-state geotherm (Chapman, 1986).
232
Chapter 6
There are several processes that may affect the temperature distribution locally or regionally, including erosion and sedimentation (thermal blanketing), among others. In general, it is the most recent thermal event that controls the thermal structure of the continental lithosphere.
3.2. RHEOLOGY MODELS OF THE EASTERN CORDILLERA THROUGH TIME 3.2.1. Rheology m odels
Sarmiento et al. (in prep) forward modelled the evolution of rheological 1-D profiles durin g the Mesozoic extension basin formation and Cenozoic basin inversion along a cross section of the E C (Fig. 6.13, see also Fig. 3.1.), using the stratigraphic record to calculate tectonic subsidence curves and lithospheric cr ustal and su bcrustal stretching events (see Chapter 2). The forward modelling ap proach is based on lithospheric stretching assumptions (McKenzie, 1978; Royden and Keen, 1980). For the Cenozoic basin inversion episodes we used the present-day topography and the Palaeogene palaeotopography estimated from flexural models (see Chapter 3, Table 3.2a/b), as well as the Neogene erosion as observed (see Chapter 4), in order to estimate compressional lithospheric factors (“stretching” crustal and subcrustal lithospheric factors β < 1, δ< 1). Compressional lithospheric factors also were independently calculated from: (1) Comparison of present-day crustal thickness from gravity data (Salvador 19 91,Fig. 6.2) and values estimated from basin lithosphere stretch ing models at the end of Cretaceous (see Chapter 2, Fig. 2.35); (2) Crustal shortening amounts due to N eogene compressional deformation calculated from r egional structural balanced cross-sections of the EC (see Chapter 5). For the continental lithosphere a model with a layer of sediments, an up per crust and a lower crust overlying a half-space mantle was adopted. The surface and base of the lithosphere are treated as fixed temperature boundaries during the basin extension/inversion history. Initial temperatures were derived from a steady-state geotherm, assuming a surface temperature of 0°C (although a real surface temperature value is ~ 25°C, the error of this assumption is <1% and affects only the ductile temperature dependent lithosphere layers), and a melting point temperature of the lithosphere of 1333 °C (these are standard values used in literature). Sarmiento et al. (in prep) applied a simple onedimensional kinematic forward model, which calculates the thermal evolution of the lithosphere using the crustal and subcrustal stretching lithospheric factors constrained by tectonic subsidence data (see Chapter 2). This model can be used to calculate the evolving geotherm resulting from thermal equilibration after stretching events and to calculate rheological strength evolutions (Van Wees and Beekman, 2000). For the inversion episodes we applied lithospheric compression by mea ns of ‘stretching’ events values less than 1 as already described. Material parameters used are listed in table 6.1. Sarmiento et al. (in prep) adopted the values of material properties from Van Wees (1994 ) and calculated rheological profile evolution using three different material parameters (listed by Ok aya et al ., 1996) for power-law creep rheologies taking into account a decrease in silica content with depth: (1) Dry quartzite for the upper crust, wet quartz-diorite for the lower crust and dry dunite for the mantle; (2) Wet quartzite for the upper crust, wet quartz diorite for the lower crust and wet dunite for the mantle, and (3) Dry granite for the upper crust, dry diabase for the lower crust and dry olivine for the mantle. However the results presented are only for the first rheology. Dry olivine rheology is combined with low-temperature plasticity (eq. 6.2b). We used a strain rate of 10 -15 s -1, corresponding to the deformation rates induced from the forward basin modelling. For the relatively fast Neogene Andean deformation we used a strain rate of 10 -14 s -1. For rheological strength calculations in the brittle domain we assumed hydrostatic pore fluid pressure ( λ ≈ 0.4) and µ = 0.75 (Ranalli, 1995). For the initial crustal thickness we adopted a value of 35 km (which is the value of the stable cratonic South American Moho depth in the eastern LLA area according to Salvador, 1991) and an initial lithospheric thickness value of 120 km. For the numerical integration of the heat conduction equation in one dimension a finite difference technique was used. The grid spacing of the 1-D vertical mesh is about 250 m. The model allows the incorporation of finite and multiple stretching/compression phases and includes thermal effects of heat production and sediment fill.
233
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
10
1400
Shortening-Thickening ratio Length initial/lenght Basin final compartment 1.38 Paleogene 1.58 Neogene Villa de Leiva Rheology model
12
s d e n A d a i r M e
11
Cucuta
7
2 1.35
1.00 1.37
basin
200 Km
Bucaramanga
Tablazo Casabe-199 Infantas-1613
A
Barinas
100
0
1.00
1300
Maracaibo Basin
Serrania de Perija
Arauca
Cascajales-1
1 1200
Cocuy
Medellin Cimitarra
ra lle i rd o C
1100
3
9
1.04 1.36
Velez Arcabu co Medios Tibasosa 1
Tunja
4
Manizales
1.01 1.72
10
Yopal
Pajarito
Inverted Cocuy Agua zul sub-basin La Maria
1.03 Suesca-1 1.44
Quipile
1000
Apul o
Planas-1
1.03
5 1.59
Ibague Q El Cobre
90 0
1.38 6 1.58
l a r t e n C
1.00 1.00
INVERTED BASIN COMPARTMENTS
Caqueza
8 1.01
1. 2. 3. 4. 5. 6.
1.37 Villavicencio
Q Olini Prado
CENTRAL CORDILLERA MIDDLE MAGDALENA VALLEY TABLAZO SUB-BASIN CUNDINAMARCA SUB-BASIN SOUTHERN CUNDINAMARCA SUB-BASIN UPPER MAGDALENA VALLEY
7. SANTANDER FLORESTA MASSIF Neiva
a e n e d r a a i n c a a M r r e a S L
80 0
800
8. S CUNDINAMARCA AND W COCUY SUB-BASINS 9. COCUY SUB-BASIN
900
1000
10. LLANOS ORIENTALES 11. MARACAIBO (CATATUMBO) SUB-BASIN Guyana Shield 12. MERIDA ANDES
1100
1200
1400
1300
1500
Figure 6.13: Location map of the regional cross-section of the EC and the lithosphere rheological models shown in figures 6.14, 6.15 a nd 6.18. Note the shortening-thickening ratios calculated for each inverted basin compartment from rheological models for the Palaeogene, and from structural balanced cross-sections for the Neogene. Thermal expansion coefficient
Layer
Initial thickness (km)
Density (Kg m-3)
Conductivity (W m-1 oC-1)
Specific heat (J kg-1 oC-1)
Heat production (ìW m-3)
Sediments
0
2500
2.6
900
1.66
3.4x10 -5
Upper crust
17.5
2800
2.6
1190
1.66
3.4x10 -5
Lower crust
17.5
2800
2.6
1190
0.5
3.4x10-5
3300
3.1
1190
0.0
3.4x10-5
Mantle
Table 6.1. Parameters used to calculate the rheological models of the lithosphere of the EC.
234
Rheology 1 (rock analogue) Quartzite (dry) Quartzite (dry) Quartzdiorite (wet) Olivine (dry)
Chapter 6 3.2.2. L ocal isostasy vs r egion al isostasy effects
In order to study local isostasy vs regional isostasy effects Sarmiento et al. (in prep) performed the following steps: (1) We modelled in 1-D the lithosph ere stretching/shortening evolution at each location based on the observed sedimentary record (see Chapters 2 to 4) and topography history inferred from flexural models (see Chapters 3 and 4). (2) We compared the stretching/shortening calculated with the models with indepe ndent stretching/shortening estimations based on crustal thickness from gravity (Sa lvador, 1991; Fig. 6.2) and structural balanced cross-section interpretations (se e Chapter 5). The 1-D model results suggest excessive thickening of the crust in disagreemen t with gravity interpretations. The difference between the 1-D model results (which imply local isostatic compensation) and independent obser vations can be explained as the result of the flexural behavior of the lithosphere (see Chapters 3 a nd 4), which imply regional isostatic compensation. (3) Although we did not apply a true flexural correction, we used the “observed” Moho depth (from Salvador, 1991) and the observed present-day topography to correct the models, the new stretching/shortening results were more in agreement with independent stretching/shortening estimations. The rheological models presented below include this correction. 3.2.3. M odel r esul ts
Figures 6.14, 6.15 a nd 6.16 sh ow the modelled depth-dependent strength of the lithosphere, the integr ated lithospher e strength and their evolution throughout the Mesozoic basin extension and Cenozoic basin inversion history of the EC . Figure 6.17 s hows the modelled heat flow at the top of the basement resulting from such tectonic history. This evolution was the result of thermal re-equilibration after basin extension and inversion. For these calculations we neglected the effect of gravitational buoyancy forces arising from lateral variations in crustal thickness and temperature (Buck, 1991), because they contribute to no more than 0.5 T/m (0.5x10 12 N/m). The integrated strength can be taken as a measure to overcome observed lithospheric deformation. In this case decrease or increase of strength values relative to initial values (representing the unaffected margins: La Maria well in the unaffected LLA area in figures 6.14 a nd 6.15) are indicative for localization, locking/widening or deformation activity. This approach is valid if deformation is homogeneously distributed in the lithosphere on the scale of the basin. Because at each location the model is 1-D we assumed local strength estimates for lithospheric deformation for the entire extension and inversion history. 3.2.4. Compar ison of li thosphere str ength estim ates fr om r heological models and f lexur al models
Results of rheological models of the EC are in partial agreement with lithosphere strength estimates from flexural mod elling. Both types of models indicate crust-mantle decoupling with a weak lower crust (see Fig 4.15). In agr eement with f lexural models calculated for Palaeogene time, rheological models at the end of Cretaceous and Palaeogene time indicated a weak lithosphere in the area of the Mesozoic extensional basin. The latter is characterised by integrated strength values lower than 12 TN/m for extension and lower than 19 TN/m for compression, and effective elastic thickness values between 1 and 5 km. Additionally, both models indicate a stronger lithosphere in the stable LLA area. Such a stronger stable lithosphere, not affected by Mesozoic stretching, is characterised by integrated strength values of 12 TN/m for extension, 19 TN/m for compression and EET values between 50 and 60 km. However, results of rheological models and flexural models for the present-day are in disagreement. Rheological models suggest that the lithosphere of the EC is stronger (integrated strength values for compression σ L > 22 TN/m) than that of the stable South American lithosphere in the eastern LLA area (σ L ≈ 22 TN/m). Flexural models suggest an opposite trend with a weaker lithosphere
235
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
END JURASSIC (142 Ma)
NW
Magdalena-Tablazo sub-basin
Paleo-Magdalena Valley 1s le
Santander-Floresta paleo-high
o c ja le
a V
A
M
z a
e
b
c s C
r
toi a
a
a
a
os oi d
cr
L
P
T
Strength scale (MPa)
M aj
bi
e
iar
as s
u
a
SE
-1000 -500
0
5 00 1 00 0 1 50 0 Compression
Tension
0 20
Moho
)
m
a
k(
14 12 Integrated 10 8 strength 6 (TN/m) 4 2 0
th
40
p e
D
60 80 100
END CRETACEOUS (65 Ma) Paleo-Magdalena Valley
Magdalena-Tablazo sub-basin -1 s el
Santander-Floresta paleo-high o
aj
a
c z
u
s a
V
r
d
C
e
ac
M
T
a
A
r
iar a
M
o io
b le
s e
bi
Llanos Orientales
toi
s
sc
a
Cocuy sub-basin ja
a P
a L
Moho
b
14 12 Integrated 10 8 strength 6 (TN/m) 4 2 0
0
100 km
Ap proxim ate hori zonta l and vertica l scal e
Figure 6.14: Depth dependent rheological models of the EC calculated for Mesozoic time: (a) End of Jurassic (142 Ma); (b) End of Cretaceous (65 Ma).
in the EC (EET=25 km, assuming a continuous elastic plate, or locally EET = 0 km assuming a broken plate under the EC) and a stronger lithosphere in the stable South American eastern LLA area (EET=50 km). Flexural models calculated for present time indicate a weak lithosphere under the EC; in contrast rheological models suggest a strong lithosphere under the EC. Two important effects not considered in the standard rheological models of the EC may explain the difference:
236
Chapter 6
EN D P AL EO G E NE ( 16.4 M a)
NW
Magdalena-Tablazo sub-ba sin
Paleo-M agdale na Valley 1s
T u n ja A x ia l Region
el co ja
Pa leo-Central Cordillera a
z C
V
le a
T
P
M
a
ja
tori s a
d
o
s a
s io
rc
e
A
Cocuy sub-basin
a
u b
sc
SE
e
ib
L lan os Orientales S tr e n g t h s c a le ( M P a ) -10 0 0 - 5 00
0
50 0 1 0 00 1 5 0 0 Compression
0
Tension 20
Moho
a
Integrated strength (TN/m)
)
m
45 40 35 30 25 20 15 10 5 0
k(
40
ht p e
D
60
80
10 0
P RE SE N T D AY Magdalena Valley
Inve rted MagdalenaT a b la z o s u b -b a s i n -1
T u n ja A x i a l Region
s
o le
s a d
c
z a
e
C
ja
M
itr T
a
bi
M
V
iar
a
s e
A
1-
o
o io
rc le
s
s
u b
a
Llanos Or ie ntales
a
c aj
InvertedCocuy sub-basin
s a
P
a L
a n la
P
Moho
Integrated strength (TN/m)
b
45 40 35 30 25 20 15 10 5 0
.
..
0
1 0 0 km
A p p rox im at e h or izo nt al an d v ert ic al s ca le
Figure 6.15: Depth dependent rheological models of the EC calculated for Cenozoic time: (a) End of Palaeogene (end of early Miocene, 16.4 Ma); (b) Present time (0 Ma). The present time models illustrated here were calculated assuming local isostasy. Theoretical Moho geometry inferred from these models may suggest mantle lithosphere subduction. Compare with the present time models calculated with the “observed” Moho depth (from Salvador, 1991) interpreted to represent flexural isostasy.
237
(
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure i Compression 45
40
35
30
25
20
h
15
10
L
5
Ag e (Ma)
a
0 30 0
25 0
20 0
15 0
10 0
50
0
50
0
Exte n s io n 40
35
30
Cascajales-1 Middle Mag dalena Valley
MV
25
Velez Ar ca buco Los M edios
Magdalena-Tab lazo inver ted s ub-basin
EC
Tibasosa Pajarito
Foresta Massif
La Mari a -1
Llanos B asin
20
h
Cocuy inverted s ub-basin LL A
L
15
10
5
0 30 0
25 0
20 0
b
15 0
10 0
Ag e ( M a )
Figure 6.16: Modelled lithosphere integrated strength evolution at several locations along the regional cross-section of the EC shown in Fig. 6.13. (a) Lithosphere strength to compression; (b) Lithosphere strength to extension.
238
Chapter 6
(1) The presence of crustal discontinuities that weaken the lithosphere and play a crucial role in strain localisation (Van Wees, 1994; Van Wees and Stephenson, 1995; Ziegler et al ., 1995). Deep reaching pre-existing crustal discontinuities, such as lithological inhomogeneities associated with ancient suture zones and faults, cause significant weakening of the upper crust. Such discontinuities are apparently characterised by a reduced friction angle particularly in the presence of fluids (Sibson, 1995; Van Wees, 1994; Ziegler et al ., 1995; Sibson, 1995). During the Neogene the complete basin inversion of the EC probably linked weak upper crust faults with the weak lower crust, reducing the integrated lithosphere strength. However, predictions from rheological models of the EC suggest that present-day deformation would be restricted mainly close to the borders of the EC, which is in agreement with shallow upper crust al seismicity, which is concentrated along the eastern and western borders of the E C ( Figs. 6.5 and 6.18). Upper crustal seismicity is a general feature observed in many other areas ( e.g. European l ithosphere, Cloetingh and Burov, 1996). Intraplate seismicity is essentially restricted to the upper crust providing additional support to the notion of upper crust-lithospheric mantle decoupling, as suggested by Cloetingh and Burov (1996) for the European lithosphere. 80
70
60
50
40 300
250
200
150
100
50
0
Age (Ma ) Cascajales-1 Middle Mag dalena Val ley
MV
Velez Arc a buco Los M edios
Magdalena-Tab lazo inver ted s ub-basin
EC
Tibas osa Pajarito
Foresta M assif Cocuy inver ted sub-basin
La Mari a -1
Llanos B asin
LL A
Figure 6.17: Modelled heat flow at the top of the basement at several locations along the regional cross-section of the EC shown in Fig. 6.13. Input data were the tectonic subsidence curves shown in Fig. 3.16. The heat conduction from the asthenosphere, heat generation from the crust and sediments blanketing effects, but it does not include heat generation from the sedimentary rocks filling the basin or convective heat transport. Differences between these modelled results and actual measurements may in part be due to these two effects, as well as additional effects discussed in the text.
(2)
Anomalous heat input to the lithosphere additional to the thermal effects included in the rheological models which are: (a) Thermal conduction; (b) Heat radioactive generation in the upper crust; and (c) Thermal blanketing effect of the sedimentary cover. The local presence of Neogene volcanic rocks in the northern part of the Tunja axial re gion of the EC and s outhwest Floresta Massif (Renzoni et al ., 1967; Romero and Rincón, 1990, Figs. 6.3 a nd 6.4), elevated geothermal gradients ( Fig. 6.10) as well as several hot springs in the central and eastern parts of the EC (Forero, 1958) suggest that such additional heat input exists. Modelled strength profiles are inconsistent with the observed seismicit y (Fig. 6.18). The strong mantle lithosphere predicted by the rheological models is characterised by a practical absence of seismicity. Standard rheological models of the EC for the present time are inconsistent with flexural modelling
239
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure Magdalena Valley Basin
Llanos Basin
Easter n C ordillera Paipa - Iza volcanic rocks
NW
SE
1 0 0 km
a
2 0 0 km
0
1 0 0 km
Aqui ta in e Ba sin
2 0 0 km
Ebro Basin
Pyrenees
LATE YPRESIAN Ar ize
NP F
B ói xol s
Mont sec
Nogueres Om
Rialp
c Subducted lower crust
Magdalena Valley Basin
Easter n C ordillera
0
50 Km
Llanos Basin
b Figure 6. 18: (a) Depth dependent rheological models of the EC at present time (0 Ma) along a regional section of the EC showing the structure (after Colletta et al ., 1990) and observed seismicity (after Taboada et al., 1999, 2000). The schematic deep structure of the EC is the hypothesis proposed here. (b) Numerical geodynamic model of a collisional orogen (from Beaumont et al ., 2000). (c) Crustal cross-section of the Pyrenees (from Beaumont et al ., 2000). Compare the hypothetical structure of the EC (a) with the geodynamic numerical model of a collision orogen (b) and with the structure of the Pyrenees. See section 6 for discussion.
240
Chapter 6
results, local volcanic rocks and hot springs in the EC and the observed intermediate seismicity. This point will be discussed in section 6.6o f this chapter. 3.2.5. M odels of evolu tion of l ith osphere rheology and basin extension and in ver sion in the EC
At the end of the Jurassic the reduced integrated strength values for extension ( σ L < 12 TN/m) indicate a weak lithosphere under the Magdalena-Tablazo sub-basin. Clearly this weak lithosphere was the result of heating and thinning of the lithosphere as a result of Triassic and Jurassic rifting in this area. Models predict that further deformation and strain localization would occur in this region. In contrast, the models suggest that the unaffected lithosphere east of the Magdalena-Tablazo sub-basin (σ L ≈ 12 TN/m) would not be prone to further deformation. However, Early Cretaceous rifting did occur in this eastern region, generating the Cocuy sub-basin (se e Chapter 2). In the Magdalena-Tablazo sub-basin further rifting occurred during Early Cretaceous ti me (see Chapter 2) in agreement with model predictions, though in an opposite sense to the predicted rifting localization in the western Magdalena-Tablazo sub-basin. Evidently, the lithospheric strength models are not completely in agreement with observed deformation during Early Cretaceous time. Because the model includes thermal blanketing effects due to sedimentation, this effect cannot be used to explain the difference between model prediction and observation. Following Van Wees (1994) the difference may be due to weak fault/suture zones with considerable lower strength values than those predicted from bulk properties of the lithosphere. Development of the Cocuy sub-basin during the Early Cretaceous probably was due to reactivation of a weak normal fault zone bounding a Palaeozoic rift (Hossack et al ., 1999). This weak zone also represents the Palaeozoic (Toussaint, 1993) or Precambrian (EtayoSerna et al ., 1983) Guaicáramo frontier along which the Chibcha Terrane was accreted (Etayo-Serna et al , 1983; Toussaint, 1993). During Early Cretaceous a wide rift system developed with two major rift sub-basins, he Magdalena-Tablazo and the Cocuy sub-basins, with the less subsiding Santander-Floresta palaeo-high between them (see Chapter 2). Models suggest that at the end of Cretaceous thinning and heating of the lithosphere produced by rifting clearly resulted in reduced integrated strength values for extension ( σ L < 12 TN/m) in the wide rift system. The two sub-basins were characterized by strength values for extension lower than 10 TN/m. The Santander-Floresta palaeo-high was characterized by strength values for extension intermediate between those of unaffected lithosphere of the LLA and those of the rifted sub-basins (10 TN/m <σ L < 12 TN/m). In accordance to model predictions Palaeogene deformation mainly affected the weakest lithosphere of the two-rifted sub-basins that were slightly inverted, generating some local palaeo-topography (se e Chapter 3). Local sedimentary basins were preserved in the MV, Tunja Axial region (former S antander-Floresta palaeo-high) and LLA (see Chapter 3). Rheological models suggest that at the end of the Palaeogene (end of Early Miocene) lithosphere strength in the former extensional basin had increased due to thermal re-equilibration cooling and slight basin inversion. However integrated strength values for compression in that region (σ L < 19 TN/m) still were lower that those of unaffected lithosphere ( σ L ≈19 TN/m). In accordance with these model predictions further compressional deformation affected weak lithosphere ( σ L < 19 TN/m). During the Neogene the original extensional basin was inverted (se e Chapter 4) in accordance with rheological model predictions. Deformation affected a wide belt of th e EC. Rheological models calculated at the present time suggest a strong lithosphere in the area of the EC characterized by integrated strength values for compression ( σ L > 22 TN/m and even σ L > 35 TN/m) greater than those of unaffected LLA lithosphere ( σ L ≈ 22 TN/m). Only the deepest proximal depocenters of the Neogene MV and LLA foreland basins are characterized by integrated strength values lower that those of unaffected lithosphere ( σ L < 19 TN/m). This result probably is related to thermal blanketing effect due to thick sediments in these proximal foreland depocenters. However flexural models suggest a weak lithosphere in the EC (see Chapter 4). As mentioned before probably the presence of crustal discontinuities and anomalous heat input explain the difference. Shallow seismicity suggests that at the present-day, brittle deformation is mainly active at its frontal eastern and western borders. Probably most of the present-day brittle deformation is concentrated by transpressional reactivation of the
241
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
external fault systems that bounded the Mesozoic extensional basin and their short-cuts (s ee Chapters 4 and 5). If these rheological models were reliable they would allow us to predict that furthe r deformation of the EC is locked. Therefore further deformation would rather be restricted to the proximal depocenters of the MV and LLA foreland basins close to the borders of the EC. These predictions seem to be in agreement with present-day shallow upper crusta l seismicity, which mainly is concentrated along the eastern and western fronts of the EC ( Figs. 6.3 a nd 6.5). During Palaeogene (see Chapter 3) and Neogene (see Chapter 4) time, deformation affected a wide belt including most of the EC. Localisation of deformation in the EC can be attributed in part to the occurrence of rheologically permanent weak zones and additional heat input not considered in the standard rheological models.
4. EVOLUTION OF STRESSES AFFECTING THE LITHOSPHERE OF THE EASTERN CORDILLERA In addition to lithosphere rheology the second key parameter controlling lithosphere dynamics is stress. I review here the possible plate-tectonic-related stress evolution that affected the lithosphere of the study area.
4.1. MESOZOIC The inferred Mesozoic stretching events (see Chapter 2) probably were initially produced by tensional/transtensional stresses associated with the separation of South A merica and North America during early Mesozoic time, and later by backarc extension ( Figs. 2.3 and 2.4) as suggested by their temporal correlation with gaps in magmatic activity or redu ced activity in the Ce ntral Cordillera ( Fig. 2.25). If the plutonic belts of the Central Cordillera were developed as subduction -related mag matic arcs during Mesozoic times, as suggested by Aspden et al ., (1987), the extensional basins behind them may be interpreted as backarc basins. Where the rate of subduction exceeds the rate of convergence, extensional deformation occurs in the overriding plate. Gravitational forces acting on the subducting slab give rise to seaward retreat of the subduction boundary (Royden, 1993a, b). Trench suction forces trigger extension in the overriding plate which subsequently fills the space created by the retreating slab (Shemenda, 1993). Extension in the upper plate may also be produced by horizontal mantle flow acting on top of the subducted plate, or to pulling out of the subducted plate from under the overriding plate (Shemenda, 1993). Backarc extension is primarily driven by a combination of plate boundary forces that cause subduction and body forces that act on the sinking slab. Important strike-slip components have also been suggested by Aspden et al . (1987). According to Aspden et al. (1987) the Triassic magmatic belt was controlled by strike-slip faults. Aspden et al ., (1987) suggested that oblique convergence and an offset in the subduction zone along a major NE-SW transform fault could account for the notable absence of Cretaceous plutonism in southern Colombia and Ecuador.
4.2. CENOZOIC Build up of intraplate compressional/transpressional stresses in the NW margin of South America during latest Cretaceous-Palaeogene (see Chapter 3) could have been favoured by a subduction impediment caused by the arrival of more buoyant oceanic crust, such as an oceanic plateau (c.f. Nivia, 1987; Kerr et al., 1997; Sinton et al., 1998). In Colombia all plate-tectonic interpretations (e.g. Pindell and Barret, 1990; Pindell, 1993; Pindell and Erikson, 1993; Pindell and Tabut, 1995) propose collision of the Caribbean with northwestern South America. Collision was oblique and diachronous, becoming younger northward (Pindell and Erikson, 1993; Pindell and Tabut, 1995). Therefore plate-tectonic history suggests that Palaeogene basin inversion was collision-related, probably associated with the right-lateral transpressional deformation that led to pre-Andean orogeny in the Central Cordillera. Transpressionally deformed grabens are also associated with zones of major
242
Chapter 6
wrench faulting (Ziegler et al., 1995). In Colombia some right-lateral strike-slip faults ( e.g. Palestina Fault, Irving, 1971) could have been active during the Palaeogene. Compression in the region behind a magmatic arc is associated with Andean type orogens and occurs during periods of increased convergence rates between the subducting and overriding plates (Ziegler et al., 1998). Inversion of rift tensional hanging-wall basins located behind a magmatic arc is the result of acceleration of convergence rates between the colliding plates, their increased mechanical coupling and the transmission of compressional stresses into the backarc domain of the overriding plate (Uyeda and McCabe, 1983; Ziegler, 1993 in Ziegler et al., 1998). Rates of the Caribbean-South America plate convergence changed during the Cenozoic. Periods of development of compressional structures seem to correlate with times of high convergence rate, particularly dur ing the Eocene (Daly, 1989) and during Late Miocene-Pliocene Andean orogeny (Cooper et al., 1995; see Chapter 4). During Late Oligocene-Early Miocene development of compressional/transpressional structures was probably associated to the rupture of the Farallon Plate into the Cocos and Nazca plates about 25 Ma ago (Wortel and Cloetingh, 1981; Duncan and Hardgraves, 1984). In addition Neogene basin inversion and Andean orogeny has been correlated to the collision of the Chocó Block (Cuna Terrane according to terminology of Toussaint, 1993, 1995a, b) with the north western margin of South America (DuqueCaro, 1990, Cooper et al., 1995; Taboada et al ., 1999, 2000; Fig. 6.8).
5. MESOZOIC-CENOZOIC TECTONIC EVOLUTION OF THE EASTERN CORDILLERA 5.1. MESOZOIC EXTENSIONAL BASIN FORMATION During the Triassic and Jurassic (see Chapter 2) tensional/transtensional stresses probably initially related to the break-up of Pangea, and later to backarc extension, resulted in lithosphere stretching that generated narrow rifts (< 150 km wide), located in the places of the present-day MV and the western flank of the EC (Magdalena-Tablazo sub-basin). Repeated stretching events in the same area suggest that strain localisation affected a weak lithosphere. Thermal heating associated with stretching and reactivation of crusta l discontinuities could have contributed to such a strain localisation. During the Early Cretaceous (see Chapter 2) tensional/transtensional stresses probably related to backarc extension produced new episodes of lithosphere stretching, and generated a wide (> 180 km wide) system of asymmetric half-rift basins. These stretching events generated an asymmetric half-rift basin with a depocenter located along the palaeo-eastern flank of the EC and a major normal fault system in its eastern border (Cocuy sub-basin). A system of horst blocks was also located in the area of the Santander and Floresta massifs. A less developed second order half-rift developed at the southwestern flank of the Cordillera (Magdalena-Tablazo sub-basin). Stretching in the MagdalenaTablazo sub-basin can be related to previously weakened lithosphere, but development of the new Cocuy cub-basin only can be explained by reactivation of older crustal discontinuities, probably the pre-Mesozoic Guaicáramo palaeo-fault. Two layered stretching models with stretching factors of up to 1.66 for the crust and up to 3.49 for the subcrustal lithosphere suggest that some decoupling occurred between the crust and subcrustal lithosphere, or that an increased thermal thinning affected the mantle lithosphere. In places of maximum crustal stretching (greater than 1.4), small mafic intrusions were emplaced during the Cretaceous (Fabre and Delaloye, 1983; Moreno and Concha, 1993). During the Late Cretaceous, subsidence was produced by thermal relaxation of the lithosphere.
5.2. CENOZOIC TRANSPRESSIONAL INVERSION OF MESOZOIC EXTENSIONAL BASINS During the Palaeogene (see C hapter 3) compressional/transpresional stresses, probably initially related to collision of the Sou th American plate margin with a relatively buoyant oceanic plateau and later related to acceleration of rates of plate convergence and collision to the Chocó Block during Neogene time, affected a weakened lithosphere below the former extensional basins. This resulted in inversion episodes. Incipient basin inversion affected the we ak lithosphere of the former Magdalena-Tablazo and Cocuy sub-basins. Later, during the Neogene (se e Chapter 4), the EC resulted from the inversion of the Mesozoic extensional basins ( e.g. Colletta et al ., 1990; Cooper et al ., 1995;
243
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Casero et al ., 1995, 1997) favoured by a previously weakened lithosphere. Basin inversion explains the coincidence in location of the former extensional basins and the EC. Results of tectonic subsidence analysis (see Chapter 2) and flexural modelling (see Chapters 3 and 4) agree with this interpretation. These results suggest that co ntractional reactivation of Mesozoic exten sional faults was initiated during Palaeogene times, leading to an initial slight basin inversion; then during Neogene time the compressional deformation increased in rate and magnitude leading to a complete inversion of the original extensional basin. Scarce fission-track data and the results of flexural models suggest that moderate (< 1000 m but mostly ≤ 200 m) and local surface-uplift and exhumation of blocks, mostly associated to the master norma l fault systems delimiting the Mesozoic extensional basins, was initiated during Palaeogene times (see Chapter 3). However, only during Neogene times, increased surfaceuplift (> 1000 m) affected the former Lower Cretaceous depocenters adjacent to the master normal fault systems delimiting the basin (see Chapter 4). Complete inversion of the master normal fault systems delimiting the former extensional basins probably lead to “extrusion” of the sedimentary fill of halfgraben basins that now form the eastern and weste rn flanks of the EC. Regional structural balanced cross-section interpretations ( e.g. Colletta et al ., 1990, Fig. 6.5; Cooper et al ., 1995, Fig. 3.37 ) suggest that maximum dip-slip displacement and shorteni ng occurred on the thrust fau lts connected with inverted Mesozoic normal faults that now approximately delimit the Lower Cretaceous outcrops of both flanks of the EC. In these uplifted flanks, Lower Cretaceous or older exposed rocks indicate that all Late Cretaceous and younger sediments have been eroded. However only in the Santander, Quetame and Garzón massifs all the sedimentary cover has been removed by erosion. Proximity of these massifs to important strike-slip faults suggests that inversion was associated with important strike-slip motions. Inversion of normal faults is only possible if there is a strike-slip component (Mandl, 1988). Preservation of the Palaeogene sedimentary record in the axial Bogotá-Tunja zone of the EC, as well as of the Neogene sedimentary record of the Sabana de Bogotá area suggest that this axial region remained low during all Miocene time and was surface-uplifted only during Pliocene-Pleistocene time. Flexural model results also support this interpretation. The exceptional palynological record of the Neogene sediments of the Sabana de Bogotá area documents this uplift history which occurred between 5 and 3 Ma b.p. (Helmens, 1990; Andriessen et al ., 1993).
6. COMPARISON OF THE EASTERN CORDILLERA WITH SIMILAR MOUNTAIN BELTS, ANALOGUE AND NUMERICAL MODELLING EXPERIMENTS AND HYPOTHESES ABOUT THE DEEP STRUCTURE OF THE EASTERN CORDILLERA In this section I highlight some structural features of the EC that are common to other mountain belts, and I make comparisons with similar mountain belts, analogue and numerical models from literature in order to propose some hypotheses about the deep structure of the EC.
6.1. ASYMMETRY OF THE EASTERN CORDILLERA Opposite vergence of the EC is not symmetric (Colletta et al ., 1990). Regional balanced crosssections perpendicular to the Cordill era at the latitude of Tunja ( e.g. Colletta et al ., 1990 Figs 5.4 a nd 6.18; Cooper et al ., 1995, Fig. 3.37) show: (1) A wide zone of east-verging thrusts located in the Tunja region and the eastern flank of the Cordillera. A large amount of eastward-directed shortening occurred in this zone. (2) A narrow zone of west-verging faults at the western front of the mountain range. A moderated amount of westward-directed shortening occurred in this zone. (3) A central zone of shallow pop-up basement (Vélez-Arcabuco anticlinorium) bordered by opposite vergence reverse faults located west of the axial Tunja region (Colletta et al ., 1990).
244
Chapter 6
6.2. COMPARISON AND SIMILARITIES WITH THE PYRENEES AND THE MÉRIDA ANDES OF VENEZUELA It is interesting to note some r emarkable similarities between the EC ( Figs 3.1, 6.14 a nd 6.15) and the Central Pyrenees ( Fig. 6.18 and 6.19). In the following comparison information ab out the Pyrenees have been extracted from Chouk roune et al . (1989), Rour e et al . (1989) and Beamount et al . (2000). Both mountain ranges show the following features (compare Figs. 6.5, 6.18 a nd 6.19): (1) An asymmetrical tectonic double-wedge structural geometry in cr oss-section. The more developed wedge (eastern flank of the EC and southern flank of the Pyrenees) is wider than the other (western flank of the EC and northern flank of the Pyrenees). Correspondingly, displacement and cumulative shortening is also greater in this wedge. Thrust transport direction in each wedge is toward the external part of the mountain range. Between the two external wedges there is an antiformal structure with shallow basement. In the EC this antiformal structure is the antiformal pop up in the Vélez-Arcabuco anticlinorium. In the Pyrenees the corresponding is an antiformal stack where basement rocks crop out. (2) Areas of outcropping basement occur at places where the sedimentary cover has been eroded. These areas are bounded by strike-slip faults. In the EC the Bucaramanga-Santa Marta and Altamira faults bound the Santander Massif and Garzón-Quetame Massif, respectively. In the Pyrenees the basement antiformal stack is bounded by the North Pyrenean Fault. (3) Two foreland basins are preserved one in each side of the double-wedge. The wider foreland basin is located adjacent to the more developed wedge. In the EC close to the more evolved eastern wedge there is the wider LLA Basin. In the Pyrenees close to the more evolved southern wedge there is the wider Ebro Basin. (4) A similar geological history: In both areas, extensional basins were developed during the Early Cretaceous related to extensional faults presently located at both sides of the orogenic doublewedge. Marine sediments accumulated in the hanging-wall strongly subsiding blocks of these extensional faults. There was a first stage of convergence and inversion that started to develop foreland basins. During Paleocene time sedimentation changed from marine to continental in these foreland basins (although in the EC case this is only applicable to the MV). Comparison of the regional cross-sections of the EC published by Colletta et al . (1990; Figs. 5.4 and 6.18) and Cooper et al . (1995, Fig. 3.37) with a series of balanced cross-sections of the Pyrenees showing its evolution through time (Fig 6.19; Choukroune et al ., 1989; Choukroune and ECORS team, 1989; Roure et al ., 1989; Beaumont et al , 2000) indicates that the closest similarity is between the present-day structure of the EC and the Pyrenees during late Ypresian time (early Eocene). The most remarkable similarity between the two mountain ranges is their asymmetry. However the comparison should not be extended beyond because in other aspects the EC and the Pyrenees seem to be different. Similarities between the EC and the Pyrenees (a collision orogen) indicate that the EC has some features of a collision orogen. However the EC was not formed during collision of two different continental plates as defined for collision orogens, it was formed by inversion of a Mesozoic extensional basin. True collision orogens, such as the Alps, have ophiolite suites associated with a suture formed during the closing of an oceanic basin. However because the EC was not formed by the closure of an oceanic basin it lacks such ophiolitic suture zones. The lithospheric structure of the Pyrenees has been deduced by a range of geophysical techniques (deep reflection and refraction seismic profiles, gravity, magnetotellurics, magnetic anomalies, tomography and heat flow). Based on these data and geological data, Roure et al . (1989), Roure (1998), Beaumont et al . (2000) suggested that the orogenic double wedge involves mainly upper crustal rocks.
245
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
PRABONIAN Tertiary foreland basin
Arize
Continental piggy-back basins M on ts ec
B ói xo il s
NPF
salt sedimentation 36Ma
50Ma
60Ma
80Ma
Nogueres 90km
Orri Rialp
EARLY OLIGOCENE Arize
Bóixoils
NPF
Montsec
Serres Marginals
36Ma
30Ma
Nogueres Orri
50Ma
60Ma
80Ma
120km Rialp
Watershed devide
PRESENT Aquitaine basin
Arize
Serres Bóixoils Montsec Marginals Ebro basin 20Ma
Nogueres
NPF
Orri
30Ma
36Ma
50Ma
60Ma
80Ma
Rialp 135 km
0
50Km
S
N CENOMANIAN Bóixols
NPF
Arize
Nogueres
PALEOCENE
NPF Arize
Bóixols
Om
Rialp
Nogueres
LATE YPRESIAN
Arize
NPF
Bóixols
Mesozoic foreland basin
Montsec
Nogueres Om
SerresMarginals
Rialp
Om
Pre-collision Mesozoic
Montsec
60Ma
80Ma
35km
SerresMarginals
60Ma
50Ma
80Ma
60km
Rialp
50Km
0
Subducted lower crust
Figure 6.19: Tectonic evolution of the Pyrenees illustrated a series of balanced retrodeformable sections representing the same regional section at different times. North is in the left side of the sections (from Beaumont et al ., 2000).
246
Chapter 6
According to Beaumont et al . (2000) the crust was decoupled and the lower crust, below the upper crustal double wedge, was subducted together with the lithospheric mantle into the mantle. However Roure et al. (1989) and Roure (1998) based on the presence in the tectonic wedge of lower crustal (Castillón and St. Barthelemy granulites) and upper mantle material (lherzolite of the Lherz lobe, Roure et al., 1989 and Roure, personal communication) have interpreted that the lower crust was not subducted due to its buoyancy. According to Roure (1998) crustal balance in the Pyrenees, the Alps and similar orogenic belts does not require subduction of lower crust. Instead this author proposed that lower crust has deformed in duplex with triangle geometry. A northward subduction of the Iberian mantle lithosphere under the European lithosphere has been interpreted from deep seismic crustal reflection interpretation (Roure, 1998; Beaumont et al ., 2000). In this case the subduction vergence coincides with the vergence of the thrust faults in the more developed wedge. In contrast data on crustal structure of the EC is limited to gravity and seismicity observations. In the Andes horizontal shortening observed in the brittle crust has been balanced by a progressive stacking and thickening of the ductile lower crust in the internal part of the orogen, as indicated by the configuration of the Moho beneath the or ogen (Roure, 1998). This is applicable to the Colombian EC characterised by a thick crust (Fig. 6.2). A progressive rise of isotherms beneath the Central Andes (Colombian Central Cordillera with an active magmatic arc) accounts for progressive thinning of the continental mantle lithosphere, and the anomalous high elevation of the Central Cordillera. In the EC the presence of small areas of Neogene volcanic rocks at Paipa and Iza (Romero and Rincón, 1990), elevated geothermal gradients (F ig. 6.10), Tertiary hydrothermalism (Branquet, 1999) and hot springs (Forero, 1958) may suggest a similar relatively thin mantle lithosphere. Interpretation of flexural models suggests decoupling between the upper crust and lithospheric mantle. Colletta et al . (1990) estimated a minimum of 105 km of shortening along the Tunja regional cross-section. According to these authors this shortening is mostly due to the two thrust fronts. They assumed also that the upper crust deformed independently from the lithospheric mantle. While the upper crust formed a double wedge, the mantle lithosphere was probably subducted (Colletta et al ., 1990). Taking the top of the basement as a reference they calculated the cross sectional area of structural relief (1500 km2). With this value and their shortening estimate, they calculated a décollement at a depth of 14 km below the top of the basement. Assuming 5 to 8 km of pre-Miocene sedimentary cover, these authors estimated the depth of décollement was about 19 to 22 km before deformation. The similarities of the EC with the Pyrenees may suggest that the EC has a similar deep structure to the Pyrenees. The conceptual model of the crustal architecture of a collision orogen proposed by Roure (1998, Fig. 6.20) might be applicable to the deep structure of the EC. A comparison of the structure of the EC with the Mérida Andes of Venezuela shows similarities but also important differences: (1) Both mountain ranges show a double wedge structural geometry in cross-section. The more developed wedge in the EC related to the LLA flexural basin can be compared to the northern wedge of the Mérida Andes related to the Neogene flexural basin in the Maracaibo Block (Colletta et al ., 1997). However, the southern wedge of the Mérida Andes is dominated by both north and south-verging basement-involving structures and the southern Barinas Basin hardly compares with a flexural basin (Colletta et al ., 1997). Between the external wedges there is an antiformal pop-up structure with shallow basement. This antiformal pop-up structure is the Vélez-Arcabuco anticlinorium in the EC and the pop-up structures associated with the Boconó Fault in the Mérida Andes. (2) Both mountain ranges show internal areas of outcropping basement in places where the sedimentary cover has been eroded. These areas are bounded by strike-slip faults. In the Mérida Andes of Venezuela the main strike-slip fault is the Boconó Fault. (3) Both mountain ranges have a geological history of Mesozoic extensional basin formation followed by Cenozoic inversion. (4) Both mountain ranges show an asymmetric structure. The more developed flexural basin is associated with a strong negative gravity Bouguer anomaly. (Neogene LLA foreland basin for the EC and Neogene Maracaibo foreland basin for the Mérida Andes). Based on reflection/refraction seismics and potential field geophysical data Colletta et al . (1997)
247
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
interpreted a south-dipping subduction of the infracontinental lithosphere mantle of the Maracaibo Block under the Mérida Andes. If the analogy were valid, probably it would imply a west-dipping subduction of the infracontinental lithosphere mantle of the LLA basin under the EC. u l t, n F a a e y r e n r t h P u l t ) o N a , l i n e g a F t e s l ip u b r ic r a m a n s Frontal thrusts o a l m r e s t r ik i .e . I n s B u c a ( u p
Molasse or foreland basins (i.e. Ebro Basin, Llanos Basin)
Conjugate backthrust Foreland basin (i.e. Po Plain or Aquitaine, Magdalena Valley)
Stack lower crustal duplexes L O W E R L O W
C R U
S T
E R C R U S T
Mantle Buffer
Subduction of the mantle lithosphere
Figure. 6.20: 3D crustal architecture of a collisional orogen: oblique collision and strain partitioning. Figure shows some similarities of the EC with some collision orogens (modified from Roure, 1998).
6.3. FLEXURAL MODELLING OF BROKEN PLATE AND SUBDUCTION OF THE MANTLE LITHOSPHERE Results of flexural models suggest that the LLA lithosphere is stronger than the EC lithosphere. Taking into account the geological history of the LLA compared to that of the EC, the weaker EC lithosphere is probably hotter than the LLA lithosphere. If a “continental” subduction started to develop between these two lithospheres, probably the colder South American (LLA) lithosphere would be subducted under the hotter, more buoyant Andean (EC) lithosphere. Hébrard (1985) suggested that underthrusting of the South American crust was occurring under the EC. In the Alps the more stabilised and colder European mantle lithosphere subducts below the hotter Apulian mantle lithosphere ( c.f. Roure, 1998).
6.4. COMPARISON WITH ANALOGUE MODEL EXPERIMENTS In order to infer the deep structure of the EC, Colletta et al . (1990) compared the structure of the Cordillera with analogue models developed by Malavielle (1984) and Ballard et al . (1987; Fig. 6.21). In both small-scale analogue model experiments, the asymmetry is caused by a vel ocity discontinuity along the basal décollement. The thrust distribution with respect to the velocity discontinuity is almost the same. In the Malavielle (1984) experiment the velocity discontinuity corresponds to a west-dipping subduction. The west-vergent thrusts are permanent faults, the throw of which increases with increasing shortening. The east-verging thrusts are migrating backthrusts with smaller displacement, which is opposite to the sense of maximum displacement in the EC (east-verging thrusts). These faults were successively activated during deformation and their throw is small in comparison to the permanent thrust zone. The entire model was made up of sand; thus the model has a brittle rheology with no ductile layers, which is not realistic.
248
Chapter 6
According to Colletta et al . (1990), velocity discontinuity in the Ballard et al . (1987) experiment represents an east-dipping subduction. In this case, the east-verging thrusts are migrating major faults, and the west-verging thrusts are only backthrusts with smaller displacement, which is similar to the EC. The model has interlayered silicone putty, representing the lower ductile crust, which is a better representation of the lithosphere rheology. In this model the sand layer below the silicone putty, representing the upper lithospheric mantle developed a west-dipping thrust ( Fig. 6.21). Because the lowermost layer of the model is sand, it did not represent the asthenosphere, which is not realistic, the velocity discontinuity at the bottom of the model representing an east-dipping subduction is also not realistic. The west-dipping thrust developed in the lowermost sand layer probably suggest a westdipping subduction zone developed on the uppermost mantle contrary to the interpretation of Colletta et al . (1990). However, the absence of an analogue representation of the asthenosphere in the model makes this interpretation hypothetical.
a 1
2
3
10 cm
4 5 6
S
1- 7 Order of thrust development
7
Malavieille (1984)
S Singular point
Direction of simulated subduction
b 10 cm Backthrusts
Ballard (1987)
Forethruats
S Sand
Silicone putty
Figure 6.21: Small-scale analogue models where strong asymmetry in thrust distribution has been simulated by velocity discontinuity at or close to the basal décollement: (A) Malavielle’s model (1984) simulating a west-dipping subduction. The forethrust zone is reduced but absorbs most of the throw. The backthrust zone is more widely distributed but the throw of each thrust is slight; (B) Ballard et al. (1987) model, simulating an east-dipping subduction. Due to the silicone putty, decollement forethrusts can develop and propagate, whereas backthrusts have small throws (from Colletta et al, 1990).
249
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
6.5. COMPARISON WITH NUMERICAL MODELS OF BEAUMONT ET AL . (2000) AND ELLIS AND BEAUMONT (1999) Beaumont et al . (2000) applied to the Pyrenean mountain belt a geodynamical numerical model that calculates the deformation of a crustal layer controlled by a combination of subduction of the suborogenic lithosphere and surface denudation. Results of the model are comparable to a crustal-scale restored cross-section of the Pyrenees (Beaumont et al ., 2000). The purpose of the models is to investigate the relative importance of factors controlling the development of the mountain range rather than to reproduce the details. As Beaumont et al . (2000) pointed out, although the models have been designed to investigate the evolution of the Pyrenees, the main conclusions are of general interest because the factors analysed are common to other orogens. I will summarise and discuss the applicability of these models to the Colombian EC. 6.5.1. M odel featur es
The plain-strain vertical section model crust is deformed under kinematic basal boundary conditions in which the velocity represents subduction of the mantle lithosphere and part of the lower crust ( Fig. 6.22). In the models mantle detaches from the crust and subducts beneath the stationary retro-mantle lithosphere (Beaumont et al ., 2000). Rheology of the crust in the model is rigid-perfect plastic with a brittle non-cohesive Coulomb yield criterion (Beaumont et al ., 2000). At high temperatures the deformation mechanism changes from frictional plastic-flow to thermally activated power law creep (Beaumont et al ., 2000). The brittle ductile transition is determined dynamically as part of the calculation (Beaumont et al ., 2000). Beaumont et al . (2000) assume that the minerals which control ductile creep are ‘wet’ quartz and ‘wet’ feldspar. In their models initial crustal thickness is 31 km with upper and lower crustal thicknesses of 16 and 15 km respectively. Flexural isostatic compensation of the thickened crust is modelled by the bending of two, semi-infinite elastic beams which support the crust and float on an inviscid asthenosphere (Beaumont et al ., 2000). Crust and mantle densities are typical values. Denudation of the upper surface of the model is calculated at a rate that depends on the surface elevation. Sediments are deposited in model basins in the same way with a rate proportional to depth of the basin (Beaumont et al ., 2000). Because the parameters of the models (e.g. crustal thickness, mantle and crustal densities) and tectonic histories of the Pyrenees are similar to the EC, results of this models can also be applied to the EC. 6.5.2. Resul ts
Figures 6.22 t o 6.26 sh ow the results. Beaumont et al . (2000) models P1 and P2 (Figs. 6.22 a nd 6.23) with laterally constant crustal thickness and strength produce results that are very different to the Pyrenees and the EC structure. Their models P3 and P4 (Figs. 6.24 and 6.25) include an inherited mid-crustal weak zone that is embedded in a strong upper crust. This localised weak zone represents the zone of the former Mesozoic crustal extension (Beaumont et al ., 2000). Although precursory extensional deformation of the crust and lithosphere is not included in the model and the initial crust has a constant thickness, results of the models, particularly P3, has many similarities with the Pyrenees and the EC structure. In both (P3 and P4) models the upper crust is deformed in some way independently of the lower crust: shortening in the upper crust is consumed by deformation while shortening in the lower crust is consumed by subduction. However, Roure (based on the surface occurrence of lower crustal rocks in the Pyrenees, personal communication) believes that subduction of lower continental crust is not realistic. This is also the case for the Canadian Rockies and the Andes, where all the crustal material necessary to balance the cover shortening has been stacked at the base of the crust and account for the local roots (Roure, 1998). Model P3 represents a higher degree of upper crust-lower crust-mantle coupling than model P4 with weak coupling between upper and lower crust. The assumption of strong coupling between upper
250
Chapter 6
and lower crust in some of the models ( e.g. P3, P5) is debatable in the Pyrenees: From deep seismic reflection data the lower crust and Moho have been interpreted as important decoupling horizons, with total detachment of the subcrustal mantle and stacking of ductile lower crust, resulting in uplift of overlying upper crustal units (Roure et al , 1989; Roure, personal communication). Although this aspect remains debatable, model P3, particularly in the prediction after 90-120 km of shortening, resembles the structure of the EC ( Fig. 6.5). This is in agreement with the minimum estimate of shortening of 105 km estimated by Colletta et al . (1990). The step-up shear zones developed in the model, equivalent to major thrust systems at the borders of the EC, are located just at the borders of the weak zone between them. Clearly if major normal fault systems delimiting the former extensional basin were located in this position, they would be compressionally reactivated as predicted by the model P3. It is interesting to note that the location of the shear-up zones predicted by the model coincides with major thrust systems even when the model does not include the effect of pre-existing crustal discontinuities. Beaumont, et al . (2000) also emphasised the importance of pre-existing discontinuities in the evolution of the Pyrenees. Model prediction shows the important effect of an inherited weakness zone. Also model results indicate that the shear-up zones may signify lateral changes in crustal properties and not the subduction point (Beaumont et al ., 2000). Similar to the analogue models of Malavielle (1984) and Ballard et al . (1987), the step-shear-up zone dipping in the same direction as the subducted slab is abandoned as it is advected past the subduction point, and the retro step-up shear zone, representing a backthrust, is more permanent. In model P3, after 90-120 km of shortening, the shear-up zone with the same vergence as the subduction zone has the maximum “displacement” and is adjacent to the wider foreland basin. This is comparable to the eastern thrust system of the EC adjacent to the wider LLA Foreland. The retroshear-up zone in the model is located adjacent to the narrower foreland, comparable to the MV in the case of the EC. In both the model P3 and the EC the flanks of the orogen are characterised by antiforms that have been uplifted and eroded and the middle axial region has a synform geometry less uplifted and eroded. Also in both the model P3 and the EC the topographic profile is asymmetric with maximum elevation in the wedge underthrusted by the subducted slab (eastern flank in the case of the EC). The temporal evolution of model P3 also probably is similar to the evolution of the EC. In model P5 ( Fig. 6.26), is similar to P3 except that a subduction load acts to flex the broken beams beneath the model downward. This load represents the negative buoyancy of the subducted mantle lithosphere and consequently was increased linearly with the total amount of shortening (Beaumont et al ., 2000). The effect of the subsurface load is to decrease the uplift, which preserves a basin in the axial region. It is possible to assume that a subsurface subduction load is responsible to the preservation of the Sabana de Bogotá Basin during Neogene time. If such hypothesis is true the palaeotopography predicted by flexural models without any subsurface loads is overestimated. It is interesting to note that some of the geodynamical numerical models with a mid crustal shear zone conducted by Ellis and Beaumont (1999), predict a local thickening of the lower crust. As already mentioned the assumption of strong coupling between upper and lower crust of this model is debatable: total detachment of the subcrustal mantle and ductile lower crust has been interpreted from deep seismic profiles (Roure et al ., 1989; Roure, personal communication). Further Beaumont et al . (2000) models applied to the Pyrenees explore the role of evaporite layers to explain some particular features of the Pyrenees not observed in the EC. According to the results of these models the role of evaporite layers is to create detachment levels. Shear zones developed on the tips of the inherited weakness zone connect these evaporite detachment levels, facilitating preservation of piggyback basins. Probably the evaporite levels in the Sabana de Bogotá region facilitated the development of detachment levels that transmitted deformation to the flanks of the Cordillera, preventing or retarding the surface-uplift of the axial Sabana de Bogota region. This alternative hypothesis may also explain the late preservation of the Neogene Sabana de Bogotá Basin.
251
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Figure 6.22: In this figure and the following figures, geodynamical models of the deformation of a crustal layer are controlled by a combination of subduction of the sub-orogenic lithosphere and surface denudation represented by uniform erosion at a constant rate. In these figures ∆x represents the amount of shortening applied to each model. Model P1: strong coupling between upper crust and lower crust and between lower crust and mantle (from Beaumont et al ., 2000).
An interesting feature that geodynamical numerical models reveal is that denudation of the crust at rates approaching the rock-uplift rate can cause rapid exhumation of mid and even lower crustal rocks at the surface, and create focused shear zones such as the Insubric Line in the Alps (Ellis and Beaumont, 1999). Probably this effect partly is responsible for the observed association of major strikeslip faults and basement uplifts (Santander, Garzón-Quetame massifs) in the EC. Also (Ellis and Beaumont, 1999) models suggest that denudation can maintain the focussed shear deformation. Shear thrusting deformation in the EC seems to be concentrated in the two major thrust systems in both flanks of the mountain range (rather than in the central axial zone) where denudation has been large. Also geodynamical numerical models of oblique collision suggest that when the transcurrent component of basal velocity is smaller than 1.5 times the normal compressional component, strike-slip deformation occurs simultaneously with thrusting. If the transcurrent component is greater than 1.5 times the compressional component strain is partitioned with the strike-slip component accommodated on narrow, sub-vertical “flower” structures between the two major shear-up zones representing thrust systems (Ellis and Beaumont, 1999). Regional narrow ‘flower structures” have not been described in the EC, implying that if oblique compression occurred, the transcurrent component was not greater that
252
Chapter 6
1.5 times the normal component. However the possibility of important transcurent motions cannot be ruled out, due to the lack of detailed structural studies. For example recently Branquet (1999) has described ‘flower’ structures in the eastern flank of the Cordillera. Additionally some structural interpretations such as those by ESRI and Ecopetrol, (1994), Taboada et al. (2000, their Plate 6) and Schelling (1994) show several pop-up blocks that may be interpreted as “flower” structures. If in the EC subduction of mantle lithosphere would occur, the mechanism that initiated such a subduction remains to be explained. As suggested by Roure (1998), stacking of the lower crust possibly accounts for the over-thickening of the crust and consequent uplifting due to flexural isostasy.
Figure 6.23: Geodynamical model P2: weakly coupling between upper and lower crust. Other features as explained in Fig 6.21 ( from Beaumont et al ., 2000).
253
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Figure 6.24: Geodynamical model P3: strong coupling between upper and lower crust and between lower crust and mantle. The assumption of strong coupling between upper and lower crust of this model is debatable (see te xt for discussion). This model is characterised by an inherited mid-crustal weak zone. Other features as explained in Fig 6.21 ( from Beaumont et al ., 2000).
254
Chapter 6
Figure 6.25: Geodynamical model P4: weak coupling between upper and lower crust and strong coupling between lower crust and mantle. This model also is characterised by an inherited mid-crustal weak zone. Other features as explained in Fig 6.21 (fro m Beaumont et al , 2000).
255
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
Figure 6.26: Geodynamical model P5: strong coupling between upper and lower crust and between lower crust and mantle. The assumption of strong coupling between upper and lower crust of this model is debatable (see text for discussion). An inherited mid-crustal weak zone w as applied as in previous model. This model is characterised by a subduction load L. Other features as explained in Fig 6.21. The state of the model after 120 km of shortening has some similarity with the structure of the EC if the pro-foreland basin represents the LLA Basin and the retroforeland Basin represents the MV Basin (from Beaumont et al ., 2000).
256
Chapter 6
6.6. NEOGENE VOLCANISM AND DEEP-INTERMEDIATE SEISMICITY: TWO ENIGMATIC FEATURES OF THE EASTERN CORDILLERA Two particularities of the EC, the local presence of Neogene volcanic rocks and the deepintermediate seismicity containing a small, very active zone of seimicity known as the Bucaramanga earthquake nest (Figs. 6.3, 6.4 an d 6.5) may be the expression of a single process. 6.6.1. Neogene magmati sm of P aipa I za
In the northern part of the Tunja axial zone of the Cordillera and south-west of the Floresta Massif, Neogene volcanic rocks occur between the Boyacá and Soapaga fau lts (Figs 6.3). The volcanic rocks at Iza intrude Cretaceous and Tertiary sedimentary rocks and form a s mall dome very close to the Soapaga Fault (Romero and Rincón, 1990), that probably served as a channel for upward migration of magma. According to Romero and Rincón (1990) volcanic rocks occur in a zone of Tertiary distension. The western outcrops near Paipa cover the Gúaduas, Socha, Picacho (Maastrichtian to Eocene, Hébrard, 1985) and Pliocene Tilatá formations (Renzoni et al ., 1967). Based on stratigraphic relative position, Fabre (1987) estimated the age of these volcanic rocks as Miocene to Pliocene. At Iza petrography and chemical analyses indicate that lavas are rhyolites and alkaline rhyolites (according to the Streckeisen, 1976 classification). According to the silica-alkali diagram (Hyndman, 1974 in Romero and Rincón, 1990) they are alkaline to subalkaline rocks (Romero and Rincón, 1990) with anomalously high K 2O values (7.8-12.0%), and low Na 2O values (< 0.8%; Romero and Rincón, 1990). These authors assume an original K 2O rich magma. Later intensive potassic metasomatism and strong hydrothermal alteration affected the rocks. According to Martínez, (1989 in Taboada et al ., 1999) the rhyolites are K/Rb rich and their chemical composition suggests they are linked to partial melting of the lower crust and not to oceanic subduction. At Paipa volcanic rocks are rhyolites with a large amount of sanidine indicating potassium enrichment. Hébrard (1985) documented a strong hydrothermal alteration. According to Romero and Rincón (1990), Paipa rhyolites are calc-alkaline. Detailed geochemical or isotopic studies are not available for these Neogene volcanic rocks. The existence of several hot springs in the central part and eastern flank of the EC (Forero, 1958) may suggest a belt of magmatic activity at depth. 6.6.2. Deep-i nter mediate seismi city and the Bu caram anga ear thquak e nest
The Bucaramanga earthquake nest is a zone of intense seismicity beneath the western flank of the EC, south of Buc aramanga. It is centred at 6º48’ N latitude, 73º10’ W longitude, and 161-km depth (Figs. 6.3, 6.4, 6.5, 6.7 an d 6.9, in these figures the Bucaramanga nest is located within the cluster around 150 km depth). The Bucar amanga nest is remarkable not only due to its unusually high-rate of activity and concentration, but also because of the relative paucity of earthquakes in the surrounding area (Schneider et al ., 1987). However, new data from the National Seismological Network of Colombia (NSNC) suggests that it is not an isolated phenomenon (Taboada et al ., 2000). In a period of 10 years, 75% of earthquakes of magnitude m b ≥ 3.8 for the northern Colombia came from the nest (Schneider et al ., 1987). Most seismicity is concentrated in a small volume with a diameter not greater than 20 km (Santo, 1969; Trygvason and Lawson, 1970; Dewey, 1972), and possibly 5 km (Pennington et al ., 1979, Schneider et al ., 1987). The Bucaramanga earthquake nest has the following enigmatic features in common with similar intermediate depth clusters from the Hindu Kush-Pamir region located between Pakistan and Afghanistan (Khalturin et al ., 1977): (1) Unusually high-rate of activity; (2) Small volume; (3) Paucity of seismicity in the surrounding area (Schneider et al ., 1987). However this has been questioned by Taboada et al ., (2000); (4) High values of Q. Q is a factor for transmission of seismic energy. Attenuation of seismic energy is inversely proportional to Q (Khalturin et al ., 1977; Pennington et al , 1979; Pennington, 1981); (5) Considerable variability in focal mechanism fault-plane solutions (Schneider et al ., 1987). However, Rivera (1989, in Taboada et al ., 1999) reported E-W
257
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
compression and N-S extension from a microseismic experiment around the Bucaramanga nest. Taboada et al ., (2000) confirmed E-W ó1 compression, while ó 2 and ó3 were not constrained, and tectonic stress is in between compressional and strike-slip regimes; (6) Located at complex convergent plate boundaries. Some of these features are also common to the Vrancea seismic cluster in Romania. Schneider et al . (1987) showed that the NW-SE lineation of the nest is not an artefact of the location processes (Fig. 6.3). The shape and size of the nest determined in their microearthquake study, are both accurate to 10-20% in any dimension and exhibit a predominant NW-SE lineation with no discernible dip. Based on 161 intermediate-depth earthquakes Schneider et al . (1987) found that 142 are from the Bucaramanga nest and only 12% of the total activity and a much smaller proportion of the total energy released, co mes from earthquakes that define an SEE trending alignment from 110 to 190 km depth (Fig. 6.3). Pennington et al . (1979) found that the region close to the nest is characterised by anomalously high values of Q. The average Q is the nest region is greater than 1000, which is almost three times greater than the normal Q values obtained for the upper mantle (Coral, 1985). Focal mechanisms of the Bucaramanga nest do not show correlation with their position in the nest (Schneider et al ., 1987). Based on the event-magnitude versus frequency relation observed from teleseismic events and microseismicity Schneider et al . (1987) estimated a cumulative seismic moment rate for the nest of approximately 5 x 10 23 dyn-cm/yr. From the amplitude spectra of two seismic events Coral and Sarmiento (1986) calculated the stress drop for these events applying the source model proposed by Brune (1970). This ranges between 1 and 5 bars but is based on seismic moments that are low compared to those calculated by Schneider et al . (1987). Pennington (1981) suggested that the sparse inte r mediate-depth hypocentres outside the Bucaramanga nest define a single Benioff zon e (Figs. 6.4, 6.5 a nd 6.7). According to Pennington (1981) the subducted slab is dipping approximately 20º to 25º towar d N 109ºE at depths greater than 140 km. Pennington called this the Bucaramanga segment, running from 5.2° N to 12° N in latitude and proposed that this subducted lithosphere is apparently continuous with the Caribbean Plate NW of Colombia. Seismicity associated with this subduction changes orientation north of the Bucaramanga nest. New data from the NSNC confirm a NNE (N30°E) trending subducting slab segment dipping 35 to 45 ° toward the SE, running 5.2 N to 7N in latitude along 150 km, from the north of Bogota to the Bucaramanga nest, beneath the NW flank of the EC (Taboada et al ., 2000, Fig. 6.5). This cluster was not known before the installation of the NSNC. These authors interpret this seismicity cluster as a subduction of cool and brittle infracontinental lithospheric mantle below the EC. Clearly the Bucaramanga segment of Pennington (1981) running from 5.2° N to 12° N in latitude include in its southern part the slab fragment running from 5.2 ° N to 7° N (trending N30° E and dipping 35 – 45 °) beneath the EC identified by Taboada et al ., (1999, 2000) and a northern slab between 7 ° N and 12 ° in latitude continuous with the Caribbean Plate identified by Pérez et al . (1997) (trending NE and dipping about 20 ° toward the SE, Fig. 6.7). If the Palaeo-Caribbean slab fragment below the EC is connected to the Baudó-Panamá Block as suggested by Taboada et al . (2000) it was broken from the Caribbean Plate which is subducting with a different velocity vector (Kellogg and Bonini, 1982; P érez et al ., 1997). Seismicity suggest that these two slabs have different subduction angles and geometry ( Figs. 6.7 and 6.9). The Bucaramanga earthquake nest i s located at the place where these two slabs are converging with an oblique convergence angle ( Fig. 6.9) which may explain the combination of compressional and strike-slip fault-plane solutions from the Bucaramanga nest. The boundary between these two slabs define an SEE alignment ( Fig. 6.7) parallel to the SEE alignment of the Bucaramanga nest (Fig. 6.3) identified by Schneider et al . (1987). 6.6.3. Simi lariti es between th e Bu caram anga ear thquak e nest and th e Vr ancea seismi c cluster
Association of deep-intermediate seismic clusters and magmatic activity has also been observed in the Eastern Carpathians. There, Pliocene-Quaternary alkaline basalts in the Perºani Mountains and Pliocene-Quaternary calc-alkaline high-K andesite and high-K dacite in the Harghita Mountains have
258
Chapter 6
been genetically correlated with the Vrancea seismic zone situated between 70 and 200 km depth (Gîrbacea and Frisch, 1998). The rocks are characterised by their high K 2O contents (Peccerillo and Taylor, 1976 in Gîrbacea and Frisch, 1998), reflecting some similarity with the Neogene volcanic rocks of the EC. It is interesting to note that in both the EC and in the Eastern Carpathians, both Neogene volcanism and deep-intermediate seismic clusters are located in zones of maximum curvature of the mountain belt in map view. Seismic clusters and volcanism also occur in the boundary zone of several plates or microplates (tectonic blocks) i.e. the European, Moesian and Tisia-Dacia plates in the case of the Carpathians and the South American Plate and the Andean and Maracaibo tectonic blocks (microplates?) for the Colombian Andes. Gîrbacea and Frisch (1998) have proposed a model involving slab break-off to explain the association between Neogene volcanism and the seismic Vrancea cluster. Probably a similar mechanism may ex plain the Bucaramanga earthquake nest and the Paipa-Iza Neogene volcanic rocks of the EC (Fig 6.27). 6.6.4. Th e slab break-of f model
This model has been developed to explain syn- to post-collisional magmatism (Davis and Blanckenburg, 1995), and seismic complexities observed at the Moho depth in some mountain belts (Sacks and Secor, 1990). In continental collision-type settings, the continental lithosphere is buoyant and resists subduction, while the cold dense oceanic lithosphere generates a large downward force (Davis and Blanckenburg, 1995). The net effect is extensional force acting on the transition region. The strength of the subducting plate decreases in magnitude with increasing subduction. If the deformation localises, the subducting slab began to brake in a narrow rifting mode, which contin ues until the oceanic lithosphere detaches, thus producing slab break-off (Davis and Blanckenburg, 19 95; Fig. 6.27). As a result of rifting of the subducting slab, the asthenosphere upwells into the s ubsurface rift generating a shallow thermal perturbation which may melt the metasomatised mechanical lithosphere (given its lower solidus) and provide the source for syn-collisional magmatism (Davis and Blanckenburg, 1995; Fig. 6.27). The upwelling asthenosphere could also partially melt the down-going plate (Davis and Blanckenburg, 1995). The still cold and dense fragment of oceanic lithosphere may be seismically active due to the pull of its negative buoyancy (Gîrbacea and Frisch, 1998). Davis and Blanckenburg (1995) have assessed the plausibility of this process by quantitative models and found that whether break-off will occur, and the depth at which it will occur, it is a function of temperature and hence of the subduction velocity. The model has also been applied to explain syn-orogenic magmatism in the Alps (Blanckenburg and Davies, 1995). For a subduction velocity of 1 cm/yr, breakoff could occur at depths comprised between 50 and 120 km, while at higher velocities it is still likely to occur, but at greater depths (Davis and Blanckenburg, 1995). Numerical modelling studies (Schott and Schmeling, in Willingshofer, 2000) have shown that the slab has to reach a depth of 100 – 170 km to generate negative buoyancy forces to allow for delamination and slab detachment. With increasing depth the thermal perturbation and magmatism will reduce (Davis and B lanckenburg, 1995). In the Colombian case the slab brake-off model may explain ( Fig. 6.27): (1) The Neogene Paipa-Iza volcanism, the presence of several hot springs suggesting an anomalous h eat input; (2) The reduced lithosphere strength suggested by flexural models as compared to standard rheological models; (3) The deep-intermediate seismic cluster roughly N30 E oriented and dipping towards ESE, where the Bucaramanga earthquake nest is located; and (4) The rapid surface-uplift of the EC. The presence of a subducting slab where the Bucaramanga nest is located would suggest that a relatively deep slab breakoff took place and hence could explain the reduced magmatic activity.
259
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
SE
Paipa-Iz a Volcani sm
NW
74° W
75° W
72° W
MV
EC
100 km
LLA
r
200 km
H
300 km
400 km 0
100 km
200 km
Figure 6.27: Hypothetical deep cross-section of the EC illustrating how the slab break-off model may explain the intermediate seismicity pattern, the presence of the K/Rb rich Paipa Iza volcanism. The deep subducting slab has been inferred from mantle tomographic images by Taboada et al . (2000). See text for explanation (near surface structural cross-section after Colletta et al ., 1990; seismicity after Taboada et al ., 1999, 2000).
260
Chapter 6
Probably slab break-off of an eastward subducting Caribbean-type mantle lithosphere under the EC, such as the SEE dipping slab fragment identified by Pennington (1981) and Taboada et al . (1999, 2000) occurs. If this slab represents a fragment of the Palaeo-Caribbean Plate connected to the BaudóChocó Blcok as suggested by Taboada et al . (2000), its crustal nature is similar to that of the relatively buoyant Caribbean Plate. Subduction resistance of the relatively buoyant Caribbean type crust (Burke, 1988; Kerr et al ., 1997) may have produced break-off of its denser mantle lithosp here under the Colombian Andes or a fragment of a normal denser lithosphere of the old Farallon P late (Fig 6.27). This model would explain the east-dipping slab fragment suggested by seismo logical studies (Pennington, 1981; Schneider et al ., 1987; Taboada et al ., 1999) and the Paipa-Iza volcanism. Although the SE location of the Paipa-Iza volcanic rocks relative to the seismically active portion of subducted slab under the EC seems to be inconsistent with the slab break-off model deep tomographic images interpreted by Taboada et al . (2000) suggest that a deep slab subducts eastward of the EC up to 70 ° W in longitude and up to 700 km depth. This can be a fragment of the Palaeo-Caribbean Plate as suggested by Taboada et al . (2000), or alternatively can be a fragment of a normal denser lithosphere of the old Farallon Plate. In these tomographic images the seismicity cluster where the Bucaramanga nest is located occurs just at the place where the slab changes from a shallow dip angle to a high deep angle (Fig. 6.27). The deep aseismic subducting slab east of the seismicity cluster has about 60 ° of dip, suggesting its sinking due to its negative buoyancy. This part of the slab probably was detached and is no longer connected to the surface pl ates (Fig. 6.27). The location of of the Paipa-Iza volcanic rocks may in fact, be related with an upwellin g hot asthenospheric flow at the place where the high angle subduction slab fragment is being rifted from the shallower pa rt (Fig. 6.27). Probably the hot asthenospheric upwelling generated the thermal anomaly responsible for partial melting of the lower crust under the EC and weakened its lithosphere. Partial melting of the lower crust may have generated the Paipa-Iza magmas (Martínez, 1989, in Taboada et al ., 1999). This may explain the chemical composition of the volcanic rocks which is not related to oceanic subduction (Martínez, 1989, in Taboada et al ., 1999). Magmas could have migrated upward and south-eastward along north-westdipping upper crustal faults such as the Boyacá and Soapaga faults. The location of the Iza volcanic rocks, very close to the Soapaga Fault (Romero and Rincón, 1990), supports this interpretation. An alternative hypothesis to explain the observed magmatism could be removal of the lithospheric root by delamination of the lithospheric mantle (Bird, 1979) However, this would result in extensive magmatism due to huge amounts of crustal melting, as in the Colorado Plateau. 6.6.5. Need of deep seismi c refr action and r efl ecti on data
In order to understand the deep structure of the EC, there is a need for deep seismic refraction and reflection data. All comparisons and inferences about the deep structure of the Cordillera made here are necessarily merely hypotheses. Deep seismic data could support or contradict the present hypotheses.
7. CONCLUSIONS Two major controls on lithosphere dynamics of the extensional basin formation and inversion history of the EC have been (1) lithosphere rheology, and (2) plate-related tectonic stresses. Standard rheological models and flexural models indicate that during the Mesozoic and Palaeogene a weak lithosphere resulted from lithosphere stretching in the area of Mesozoic extensional basins. However, flexural models indicate a weak lithosphere under the EC in open disagreement with standard rheological models calculated for the present time, that suggest a strong lithosphere in the EC. Two important effects not considered in the standard rheological models may explain the difference: (1) The presence of crustal discontinuities that weaken the lithosphere; and (2) The presence of an anomalous heat input into the lithosphere, as indicated by local Neogene volcanic rocks and several hot springs in the EC. During the Triassic and Jurassic tensional/transtensional stresses, probably initially related to the break-up of Pangea and later to backarc extension, produced lithosphere stretching and generated narrow rifts (< 150 km wide), located in the places of the present day MV and the western flank of the
261
Rheological Evolution of the Lithosphere of the Eastern Cordillera and Hypotheses about its Deep Structure
EC (Magdalena-Tablazo sub-basin). Repeated stretching events in the same area suggest strain localisation affecting a weak lithosphere. Thermal heating associated with stretching and reactivation of crustal discontinuities probably contributed to such a strain localisation. During the Early Cretaceous tensional/transtensional stresses probably related to backarc extension produced new episodes of lithosphere stretching and generated a wide (> 180 km wide) system of asymmetric half-rift basins, one of which with a depocenter located along the palaeo-eastern flank of the EC, with a major normal fault system in its eastern border (Cocuy sub-basin). A system of horst blocks was located in the area of the Santander and Floresta massifs. A less developed second order half-rift occurred in the place of the south-western flank of the Cordillera (Magdalena-Tablazo sub-basin). While stretching of the Magdalena-Tablazo sub-basin affected lithosphere previously weakened during earlier rifting episodes, development of the new Cocuy cub-basin only can be explained by reactivation of older crustal discontinuities, probably the pre-Mesozoic Guaicáramo palaeo-fault During the Palaeogene, compressional/transpresional stresses, probably initially related to collision of the South American plate margin with a relatively buoyant oceanic plateau and later related to acceleration of rates of plate convergence and collision to the Chocó Block during Neogene time, affected the weakened lithosphere (of the former extensional basins). This resulted in inversion episodes of the former extensional basins: During the Palaeogene, incipient basin inversion affected the weak lithosphere of the former Magdalena-Tablazo and Cocuy sub-basins, while during the Neogene, the EC resulted from the complete inversion of the Mesozoic extensional basin area ( e.g. Colletta et al ., 1990; Cooper et al ., 1995; Casero et al ., 1995, 1997). Based on surface geological data, limited geophysical data (gravity, seismicity), comparison with similar mountain belts and analogue and numerical models from the literature, it is possible to narrow the uncertainity range on the deep structure of the EC. Probably it resembles that of the Pyrenees: Lithospheric shortening has been accommodated in the upper brittle crust by development of a double vergent asymmetric wedge, while the mantle lithosphere accommodated shortening by westdipping subduction of the cooler and denser LLA mantle lithosphere under the buoyant and hotter Andean lithosphere. The lower ductile crust probably accommodated shortening by thickening. Probably the EC has been strongly affected by transpression with important transcurrent components. For example Branquet (1999) recently described ‘flower’ structures in the eastern flank of the Cordillera. Additionally some structural interpretations such as those by ESRI and Ecopetrol (1994) , Schelling (1997) and Taboada et al . (2000, their plate 6) show several pop-up blocks that may be interpreted as “flower” structures. Intermediate seismicity beneath the EC suggest that a subducted slab fragment is present below its NW margin (Taboada et al ., 1999, 2000). A small, but very active zone of intermediate-deep seismicity, the Bucaramanga earthquake nest (Schneider et al ., 1987), may result from deep oblique convergence of Palaeo-Caribbean plate fragment connected to the Panamá Block and the Caribbean Plate from the north. This may explain the NW-SE lineation of the nest (Schneider et al ., 1987). Two particularities of the EC, i.e. the local presence of Neogene volcanic rocks at Paipa and Iza, and the intermediate seismicity, may be the expression of slab break-off of the eastward-subducting Caribbeantype mantle lithosphere fragment under the EC. Subduction resistance of the relatively buoyant Caribbean type crust (Burke, 1988; Kerr et al ., 1997) may have produced break-off of its denser mantle lithosphere under the Colombian Andes or a fragment of a normal denser lithosphere of the old Farallon Plate. The deep slab suggested by tomographic images (Taboada et al ., 2000) may be a subducted slab, which is no longer connected to the surface plates. This hypothesis would explain the east-dipping slab fragment suggested by seismological studies (Pennington, 1981; Schneider et al ., 1987; Taboada et al ., 1999, 2000) and the Paipa-Iza volcanic rocks. These volcanic rocks may be related with upwelling hot asthenosphere where slab break-off occurs. Possibly partial melting of the lower crust beneath the EC may have generated rising magmas, generated a thermal anomaly under the EC and weakened its lithosphere. Deep seismic studies could offer reliable data in order to support or contradict the suggested hypotheses.
262
REFERENCES Aalto, K. R., 1971. Petrografía de las areniscas de la sección estratigráfica de Bogotá, Colombia. Geología Colombiana, 3: 7-36. Acosta, J. E., 1999. Structural evolution of the Eastern Cordillera Foothills between Nunchía and Paz de Ariporo (Colombia). Actas X Congreso Latinoamericano de Geología y VI Congreso Nacional de Geología Económica, 2: 141-148. Acosta, J., Coward, M. P. and Lonergan, L., 2000. Kinematics of the western foothills of the Colombian Eastern Cordillera and regional implications. In press, London, 28 p. Alfonso, C. A., 1985. Los episodios evaporíticos de la Formación Rosablanca. In: F. Etayo-Serna, F. and Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XVI, Ingeominas Publicación Geológica Especial 16, Bogotá, 19 p. Alfonso, C. A., 1989. Stratigraphy and regional structure of the western flank of the Cordillera Oriental, Cimitarra area, Colombia. M.Sc. Thesis Univ. of South Carolina, Columbia, SC., 117 p. Alfonso, C. A. and Ballesteros, I., 1987. Unpublished stratigraphic column, 1:5000 scale, Area 1. In: R. Ressetar, and C.E. Macellari). ESRI Tech. Rept. 88-0006, 40 p. Alfonso, C. A., Sacks, P. E. and Secor, Jr. D. T., 1989. Late Tertiary northwestard-vergent thrusting in Valle del Cauca, Colombian Andes (abs.). A.A.P.G. Bull., 73: 327. Alvarado, B. and Sarmiento, R., 1944. Informe general sobre los yacimientos de hierro, carbón, y caliza de la región de Paz de Río, Departamento de Boyacá. Servicio Geológico Nacional Informe 468, Bogotá, 132 p. Alvarez, J., 1983. Geología de la Cordillera Central y el Occidente Colombiano, y petroquímica de los intrusivos granitoides Mesocenozóicos. Ph.D. Thesis Univ. Chile. Boletin Geológico Ingeominas, Bogotá, 26(2):1-175. Alzate, J. C. and Rolón, L. F., 1996. Modelo estratigráfico secuencial para el Cretácico Inferior en las áreas vecinas a los Macizos de Floresta y Santander (Cordillera Oriental), Memoria VII Congreso Colombiano de Geología, Santafé de Bogotá, 14 p. Alzate, J. and Bueno, M., 1994. Análisis estratigráfico secuencial de las rocas cretácicas de la parte oriental del Departamento de Boyacá, Municipios de Sogamoso, Belencito y Aquitania. Tesis Pregrado Geología, Univ. Nal. de Colombia, Depto. de Geociencias, Bogotá, 90 p. Allemand, P. and Brun, J. P., 1991. Width of continental rifts and rheological layering of the lithosphere. Tectonophysics 188: 63-69. Allen, P. A. and Allen, J. R., 1990. Basin analysis, principles and applications. Blackwell Scientific Publications, London, 451 p. Allmendiger, R. W., Marret., R. A. and Cladouhos, T., 1989. Faultkin, a program for analyzing fault slip data. Amaya, S. F. and Santamaría I., 1994. Cartografía geológica y geometría estructural del sector Peralonso Tetuán al Sur de Ortega (Tolima) en el Valle Superior del Magdalena. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter VII, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 20 p. Amézquita K. F. and Montes, C., 1994. Sección geológica del Maco Buenavista: Estructura en el sector occidental del Valle Superior del Magdalena. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena.C hapter VI, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 36 p. Anderson, T. A., 1970. Geology of the lower Tertiary Gualanday Group, Upper Magdalena Valley, Colombia. Ph.D. Thesis, Princeton Univ., Princeton, 86 p. Anderson, T. A., 1972. Paleogene nonmarine Gualanday Group, Neiva basin, Colombia, and regional development of the Colombian Andes. Geol. Soc. Am. Bull. 83: 2423-2438. Andriessen, P. A. M., 1995. Fission-track analysis: principles, methodology and implications for tectono-thermal histories of sedimentary basins, orogenic belts, and continental margins. Geologie en Mijnbouw. 74, p. 1-12.
References
Andriessen, P. A. M., Helmens, K. F., Hooghiemstra, H., Riezebos, P. A. and Van der Hammen, T., 1993. Absolute chronology of the Pliocene-Quaternary sediment sequence of the Bogotá Area, Colombia. Quaternary Sci. Rev. 12: 483-503. Arango, F. and Venegas, D., 1994. Análisis Estratigráfico de una Región al Occidente de la Población de Paz de Ariporo (Casanare). Tesis Pregrado Geología, Universidad Nacional de Colombia. Santafé de Bogotá. Araujo, M., 1997. Experiencias en la exploración de frente de montaña de la Sierra de Perijá, Cuenca de Maracaibo, Venezuela. VI Simpolsio Bolivariano Exploracion Petrolera en las Cuencas Subandinas. Mem. Tomo I: 448-456. Arias, A. and Vargas, R., 1978. Geología de las planchas 86, Abrego y 97, Cáchira, Departamento de Norte de Santander, Boletín Geológico Ingeominas Bogotá, 23(2): 3-38. Aspden, J. A. and McCourt, W. J., 1986. Mesozoic Oceanic terrane in the Central Andes of Colombia. Geology, 14: 415-418. Aspden, J. A., McCourt, W. and Brook, M., 1987. Geometrical control of subduction-related magmatism: the Mesozoic and Cenozoic plutonic history of Western Colombia. Jour. Geol. Soc. London, 144: 893-905. Audemard, F., 1991. Tectonics of Western Venezuela. Ph.D. Thesis, Rice University, Houston, Texas, 243 p. Audibert, M., 1991. Déformation discontinue et rotations de blocs. Méthodes numériques de restauration. Mémoire. Documental. Centre. Armoricain Etudies. Structurales. Socles, Rennes, France. 40, 239 p. Avé Lallemant, H. G., 1997. Transpression, displacement partitioning, and exhumation in the eastern Caribbean/South American plate boundary zone. Tectonics, 16(2): 272-289. Avraham, Z., 1992. Development of asymmetric basins along continental transform faults. In: P. Ziegler (ed.), Geodynamics of rifting, Volume III, Thematic discussions. Tectonophysics, 215(1-2): 209-220. Avraham, B. Z. and Nur, A., 1987. Effects of collisions at trenches on oceanic ridges and passive margins. In: J.W.H. Monger and J. Francheteau (eds.), Circum-Pacific orogenic belts and evolution of the Pacific Ocean basin. American Geophysical Union, Geodynamics Series, v. 18, p. 9-18. Baby, P., Rivadeneira, M., Bernal, C., Christophoul, F., Dávila, F. and Galárraga, 1998. Strcutural style and timing of hydrocarbon entrapments in the Ecuadorian Oriente Basin. A.A.P.G. International conference and exhibition, abstracts. A.A.P.G. Bull 82(19): 1889. Bachu, S., Ramón, J.C., Undershultz, J.R. and Villegas, M., 1993. Evaluation of fluid and heat flow regimes in the Llanos Basin, Colombia. Ecopetrol ICP, Petrocanada Alberta Reseach Council., Edmonton, 91 p., 7 appendixes. Bachu, S., Ramón, J. C., Villegas, M. E. and Underschultz, J. R., 1995. Geothermal regime and thermal history of the Llanos Basin, Colombia. A.A.P.G. Bull., 79(1): 116-129. Ballard, J. F., Brun, J. P., Van Den Driesche, J. and Allemand, P., 1987. Propagation des chevauchements au dessus des zones de décollement: modèles expérimentaux. C.R. Acad. Sci. Paris, 305, II: 1249-1253. Ballesteros I. and Nivia A., 1985. La Formación Ritoque: Registro sedimentario de una albufera de comienzos del Cretácico. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XIV, Ingeominas Publicación Geológica Especial 16, Bogotá, 17 p. Bally, A. W., 1984. Tectogénèse et sismique réflexion. Bull. Soc. Géol. Fr., 24: 279-285. Barrero, D., 1979. Geology of the Western Cordillera, west of Buga and Roldanillo, Colombia. Ingeominas Publicación Geológica Especial 4, Bogotá, 75 p. Barrientos, A. and Torres, R., 1994. Análisis paleoambiental de la secuencia del Terciario Inferior Agua Clara Casanare (Colombia). Tesis Pregrado Geología Univ. Nacional de Colombia, Bogotá, 80 p. Barrio, C. A. and Coffield, D. Q., 1992. Late Cretaceous stratigraphy of the Upper Magdalena Basin in the Payandé-Chaparral segment (western Girardot Sub-Basin), Colombia. Jour. South Am. Earth Sci., 5(2): 132-139.
264
References
Bartels, H., 1986. El "Grupo Palmichal" y su relación estratigráfica con el Grupo Guadalupe. Tesis Pregrado Geología. Univ. Nacional de Colombia, Depto de Geociencias, Bogotá, 148 p. Bayona, G. A., García D. F. and Mora G., 1994. La Formación Saldaña: producto de la actividad de estratovolcanes continentales en un dominio de retroarco. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter I, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 21 p. Beaumont, C., 1981. Foreland basins. Geophys. J. R. Astr. Soc., 55: 471-497. Beaumont, C. and Quinlan, G., 1994. A geodynamic framework for interpreting crustal-scale seismicreflectivity patterns in compressional orogens. Geoph. Jour. Int., 116, 754-783. Beaumont, C., Fullsack, P. and Hamilton, J., 1994. Styles of crustal deformation in compressional orogens caused by subduction of the underlying lithosphere. Tectonophysics, 232, 119-132. Beaumont, C., Muñoz J. A., Hamilton, J. and Fullsack, P., 2000. Factors controlling the Alpine evolution of the central Pyrenees inferred from a comparison of observations and geodynamical models. J. G. R., 105 (B4): 8121-8145. Beekman, F., 1994. Tectonic modelling of thick-skinned compressional intraplate deformation. Ph.D. Thesis Vrije Universiteit, Amsterdam, 152p. Beltran, N. and Gallo, J., 1968. The geology of the Neiva subbasin, Upper Magdalena basin (Southern portion): 9 th Annual Field Conference Guidebook, Colombian Association Petroleum Geologists and Geophysicists., Bogotá, p. 253-274. Bessis, F., 1986. Some remarks on the study of subsidence of sedimentary basins; application to the Gulf of Lions margin (western Mediterranean). Marine and Petroleum Geol., 3(1): 37-63. Bird, P., 1979. Continental delamination and the Colorado Plateau. J. G. R., 84: 7561-7571. Blanckenburg, F. and Davies J. H., 1995. Slab breakoff: A model for syncollisional magmatism and tectonics in the Alps. Tectonics, 14(1): 120-131. Bodine, J. H., 1981. The thermomechanical properties of the oceanic lithosphere, Ph.D. Thesis, Columbia University, 332 p. Bogotá, J. and Aluja, J., 1981. Geología de la Serranía de San Lucas. Geología Norandina, Bogotá, 4: 49-55. Boinet, T., 1985. La frontiere meridionale de la plaque Caraibe aux confins Colombo-Venezuelensis (Norte de Santander, Colombie: donnes geologiques. These Diplome Docteur 3eme cycle. L'universite Pierre et Marie Curie, Paris 6, Paris, 200 p. Boinet, T. H., Bourgois, J. and Mendoza, H., 1982. Tectónica de sobrecorrimiento y sus implicaciones estructurales en el área Pamplona, Cordillera Oriental de Colombia. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 15(29): 81.97. Boinet, T., Bourgois, J., Bellon, H. and Toussaint, J. F., 1985a. Age et répartition du magmatisme Prémésozoique des Andes de Colombie. C. R. Acad. Sci. Paris. 300, II(10): 445-450. Boinet, T, Bourgois, J., Mendoza, H. and Vargas, R., 1985b. Le poinçon de Pamplona (Colombie): un jalon de la frontière méridionale de la plaque Caraï be. Bull. Soc. Géol. France, 8(1-3): 403-413. Bond, G. and Kominz, M. A., 1984. Construction of tectonic subsidence curves for the early Paleozoic miogeocline, southern Canadian Rocky Mountains: implications for subsidence mechanisms, age of break-up, and crustal thinning. Geol. Soc. Am. Bull. 95: 155-173. Bonini, W. E., Hargraves, R. B. and Shagam, R (eds.)., 1984. The Caribbean South American plate boundary and regional tectonics. Geol. Soc. Am. Memoir 162, 421 p. Botero-Arango, G., 1963. Contribución al conocimiento de la geología de la zona central de Antioquia. Anales Facultad de Minas., Medellín, 57: 1-101. Bott, M. H. P., 1992. Passive margins and their subsidence. Jour. Geol. Soc., London, 149: 805-812. Bourgois J., Azéma J., Tournon, J., Bellon, Calle B., Parra, E., Toussaint J. F., Glaçon G., Feinberg H., De Wever P. and Origlia I., 1982a. Ages et structures des complexes basiques et ultrabasiques de la facade Pacifique entre 3° N et 12° N (Colombie, Panamá et Costa-Rica).Bull. Soc. Géol. France, 7 t XXIV (3): 545-554. Bourgois, J., Calle, B., Tournon, J. and Toussaint, J. F., 1982b. The Andean ophiolitic megastructures on the Buga-Buenaventura transverse (Western Cordillera – Valle Colombia). Tectonophysics, 82: 207-229.
265
References
Bourgois J., Toussaint J. F., Gonzales H., Azéma J., Calle B., Desmet A., Murcia L. A., Acevedo, A. P., Parra E. and Tournon, J., 1987. Geological history of the Cretaceous ophiolitic complexes of Northwestern South America (Colombian Andes). Tectonophysiscs 143: 307-327. Bourgeois, O., Cobbold, P. R., Rouby, D., Thomas, J. C. and Shein, V., 1997. Least squares restoration of Tertiary thrust sheets in map view, Tajik depression, central Asia. J. Geophys. Res. 102(B12): 27.553-27.573. Bowin, C., 1976, Caribbean gravity field and plate tectonics, Spec. Pap. Geol. Soc. Am., Boulder Colorado, 169, 79 p. Branquet Y., 1999. Etude structurale et métallogénique des gisements d ′émeraude de Colombie: Contribution à l ′histoire tectono-sédimentaire de la Cordillère Orientale de Colombie. Thèse de Doctorat de l’Institut National Polytechnique de Lorraine, Specialité Geosciences, Matières premières et environement, Institut National Polytechnique de Lorraine, Centre de recherches pétrographiques et géochimiques (CRPG/CNRS), 265 p Branquet, Y., Laumonier, B., Lopes, B., Cheilletz, A., Giuliani, G. and Rueda, F., 1996. Evidences of compressive structures in the Muzo and Coscúez emerald deposits, Eastern Cordillera of Colombia. Third ISAG, St Malo, Abs., p. 675-677. Branquet, Y., Cheilletz, A., Cobbold, P. R., Baby, P., Laumonier, B. and Guiliani, G., 1999a. Andean transpressive tectonics at the eastern edge of the Cordillera Oriental, Colombia (Chivor-Guavio area). Fourth ISAG, 4-6 October 1999, Extended Abstracts, p. 103-105. Branquet, Y., Chellietz, A., Giuliani, G., Laumonier, B. and Blanco, O., 1999b. Fluidized hydrothermal breccia in dilatant faults during thrusting, the Colombian emerald deposits. In: K.J.W., McCaffrey, L., Lonergan and J., Wilkinson (eds.), Geol. Soc. Spec. Publ. Geol. Soc., London, 155: 183-195. Braun, J. and Beaumont, C., 1987. Styles of continental rifting: Results from dynamic models of lihtospheric extension, In: C. Beaumont and A.J. Tankard (eds.), Sedimentary Basins and Basin-Forming Mechanisms, Can. Soc. Petr. Geol. Memoir, 12: 241-258. Braun, J. and Beaumont, C., 1989. Contrasting styles of lithospheric extension: implications for differences between Basin and Range province and rifted continental margins. In: A.J. Tankard and H.R. Balkwill (eds.), Extensional tectonics and stratigraphy of the North Atlantic margins. A.A.P.G. Mem. 46: 53-79. Breen, N. A., 1989. Structural effect of Magdalena fan deposition in the northern Colombia convergent margin. Geology, 17:34-37. British Petroleum., 1989. Informe técnico contrato de evaluación técnica Boyacá, años 1987, 1988 y 1989, 16 Vols.; Ecopetrol Bogotá, ISN 11297, Informe 2253. Brune, J. N., 1970. Tectonic stress and the spectra of seismic shear waves from earthquakes. J. G. R., 75: 4.997-5.009. Buck, W. R., 1991. Modes of continental lithosphere extension. J. Geophys. Res. , 96: 20.161-20178. Buitrago, J., 1994. Petroleum systems of the Neiva area, Upper Magdalena Valley, Colombia. In: L. B. Magoon and W. G. Dow (eds). The petroleum system; from source to trap. A.A.P.G. Mem. 60: 483-497. Bürgl, H., 1958. El Jurásico e Infracretáceo del río Batá, Boyacá. Boletín Geológico, Ingeominas, Bogotá, 6(1-3):169-211. Bürgl, H., 1959. Apuntes sobre la estratigrafía de los alrededores de Neiva, Huila, Servicio Geológico Nacional, Informe 1318, Bogotá, 14 p. Bürgl, H., 1960. El Jurásico e Infracretáceo del Río Batá, Boyacá: Servicio Geológico Nacional, Informe 1319, Boletín Geológico, Bogotá, 1-3: 169-211. Bürgl, H., 1961a. Geología de los alrededores de Ortega, Tolima. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 8: 21-38. Bürgl, H., 1961b. Historia geológica de Colombia. Revista Academia Colombiana Ciencias Exactas Fisicas y Naturales, Bogotá, 9(43): 137-191. Bürgl, H., 1961c. Sedimentación cíclica en el geosinclinal Cretáceo de la Cordillera Oriental de Colombia. Boletín Geológico Servicio Geológico Nacional, Bogotá, 7(1-3): 85-118. Bürgl, H., 1964. El Jura-Triásico de Colombia. Boletín Geológico Servicio Geológico Nacional, Bogotá, 12(1-3):5-31. Bürgl, H., 1967. The orogenesis of the Andean system of Colombia. Tectonophysics, 4(4-6): 429-443.
266
References
Burgl, H. and Dumit-Tobon, Y., 1954. El Cretáceo Superior en la región de Girardot. Boletín Geológico Instituto Geológico Nacional, Bogotá, 2(1): 23-48. Burgl, H. and Radelli, L., 1962. Nuevas localidades fosilíferas en la Cordillera Central de Colombia. Geología Colombiana, Bogotá, 3: 133-138. Burke, K., 1988. The tectonic evolution of the Caribbean. Ann. Rev. Earth Planet. Sci. 16: 210-230. Burke, K., Cooper, K., Dewey, J., Mann, P. and Pindell, J., 1984. Caribbean tectonics and relative plate motions. In: W. Bonini, R. Hargraves, R. and Shagam, (eds.), The Caribbean South American plate boundary and regional tectonics. Geol. Soc. Am. Mem., 162: 31-63. Burov, E. B. and Diament, M., 1995. The effective elastic thickness (Te) of continental lithosphere: What does it really mean?. J. Geophys. Res. , 100(B3): 3905-3927. Burov, E. B., Lobkovsky, L.I., Cloetingh, S. and Nikishin, A.M., 1993. Continental lithosphere folding in Central Asia (Part II): Constraints from gravity and topography. Tectonophysics, 226: 73-87. Burtman, V.S. and Molnar, P., 1993. Geological and geophysical evidence for deep subduction of continental crust beneath the Pamir, Geol. Soc Am. Boulder Colorado, Spec Paper 281, 51 p. Busbey, A. B., 1986. New material of Sebecus cf. Huilensis (Crocodilia) from the Miocene La Venta Formation, Colombia. Jour. Vertebrate Paleont., 6(1): 20-27. Butler, K., 1983. Andean-type foreland deformation: structural development of the Neiva Basin, Upper Magdalena Valley, Colombia. Ecopetrol, Bogotá, Informe Geológico 1509. Butler, K. and Schamel, S., 1988. Structure along the eastern margin of the Central Cordillera, Upper Magdalena Valley, Colombia. Jour. South Am. Earth Sci., 1(1): 109-120. Butler, W.H., Harris, B.W. and Whittington, G., 1997. Interactions between deformation, magmatism and hydrothermal activity during active crustal thickening, a field example from Nanga Parbat, Pakistan Himalayas. Mineralogical Magazine, London, 61(1): 37-52. Byerlee, J. D., 1978. Friction in rocks. Pure Appl. Geophys. 116: 615-626. Cáceres, C. and Etayo-Serna, F., 1969. Bosquejo geológico de la región del Tequendama, Opúsculo guía de la Excursión Pre-Congreso. 1er Congreso Colombiano de Geología, Opúsculo, Bogotá, 23 p. Caicedo J. and Roncancio, J. H., 1994. El Grupo Gualanday como ejemplo de acumulación sintectónica en al Valle Superior del Magdalena durante el Paleoceno. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter X, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 20 p. Calvache, J. A. and Muñoz., F. A., 1984. Modelo de la corteza terrestre en un área aledaña al municipio de Cumaral. Departamento del Meta. Tesis Pregrado Geología, Univ. Nacional de Colombia. Depart. Geoc., Bogotá. 116 p. Campbell, C. J., 1968. The Santa Marta wrench fault of Colombia and its regional setting, Trans. IV Caribb. Geol. Conf., Trinidad, p. 247-262. Campbell, C. J., 1974. Colombian Andes. In: A. M. Spencer (ed.), Mesozoic-Cenozoic orogenic belts: data for orogenic studies: Edinburgh, Scottish Academic Press, p. 705-724. Campbell, J. C. and Bürgl, H., 1965. Section through the Eastern Cordillera of Colombia, South America. Geol. Soc. Am. Bull., 76: 567-590. Cardona J. and Gutíerrez Z., 1995. Estratigrafía y ambientes de depósito de la Formación Carbonera en un área al noroeste de Yopal-Casanare (Colombia). Tesis Pregrado Geología, Univ. Nacional de Colombia, Bogotá, 80 p. Cardozo E. and Ramírez C., 1985. Ambientes de depósito de la Formación Rosablanca: Área de Villa de Leiva. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XIII, Ingeominas Publicación Geológica Especial 16, Bogotá, 13 p. Cardozo, E. and Sarmiento, L. F., 1987. Unpublished stratigraphic column, 1:5000 scale, Area 3. In: R. Ressetar, and C. E. Macellari. ESRI Tech. Rept. 88-0006, 40 p. Cardozo, N. F. and Zúñiga, J., 1995. Análisis estructural de las zonas del bloque colgante (hangingwall) y bloque yacente (foot-wall) de la Falla de Soapaga entre Corrales y Paz de Río (Boyacá), Tesis Pregrado Univ. Nacional de Colombia, Empresa Colombiana de Petróleos Ecopetrol, Bogotá, 90 p. Cardozo-Puentes, E., 1989. Structural style of the central western flank of the Cordillera Oriental west of Bogotá, Colombia. M.Sc. Thesis, University of South Carolina, Columbia S.C., 146 p.
267
References
Carillo, V., 1982. La Quebrada Confines (Municipio de Cucutilla, Norte de Santander), una nueva localidad fosilífera del Paleozóico Superior en la Cordillera Oriental de Colombia. Geología Norandina, Bogotá, 5: 32-38. Carter, N. L. and Tsenn, M. C., 1987. Flow properties of continental lithosphere. Tectonophysics, 136: 27-63. Casero, P., Salel, J. F. and Rossato, A., 1995. Structural evolution of the margin and foothills belt of the Cordillera Oriental of Colombia (southwest of Cusiana). VI Congreso Colombiano del Petróleo. Abs., Bogotá, 1: 33-41 Casero, P., Salel, J. F. and Rossato, A., 1997. Multidisciplinary correlative evidences for polyphase geological evolution of the foothills of the Cordillera Oriental (Colombia). VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Cartagena, Abs., 7.Tomo II: 100-118. Castillo, J. E. and Mojica, J., 1990. Determinación de la orientación de los esfuerzos actuales a partir de elongaciones tectónicas (”breakouts”) en algunos pozos petroleros de los Llanos Orientales y del Valle Medio del Magdalena, Colombia. Geología Colombiana, Bogotá, 17: 123-132. Cazier, E. C., Hayward, A. B., Espinosa, G., Velandia, J., Mugniot, J. F. and Leel Jr W. G., 1995. Petroleum geology of the Cusiana field, Llanos Basin Foothills. Colombia, A.A.P.G. Bull., 79(10): 1444-1463. Cazier, E. C., Cooper, M. A., Eaton, S. G. and Pulham, A. J., 1997. Basin development and tectonic history of the Llanos Basin, eastern Cordillera, and Middle Magdalena Valley, Colombia: reply. A.A.P.G. Bull., 81: 1332-1335. Cediel, F., 1968. El Grupo Girón, una molasa Mesozóica de la Cordillera Oriental. Servicio Geológico Nacional Boletín Geológico, Bogotá, 16(1-3): 5-96. Cediel, F., 1982. Modelo estructural para los Llanos Orientales. Simposio Exploración Petrolera en las Cuencas Subandinas de Venezuela, Colombia, Ecuador y Perú. Colombian Society Petroleum Geologists and Geophysicists, Bogotá, 8 p. Cediel, F., 1983. El borde continental Colombiano durante el Tríasico-Jurásico y el origen de la Placa Caribe. Abstract. Mem. 10 th Caribbean Geological Conference, Ingeominas, Bogotá, p. 170. Cediel, F., Mojica, J. and Macia, C., 1981. Las formaciones Luisa, Payandé, Saldaña sus columnas estratigráficas características. Geología Norandina, Bogotá, 3: 11-19. Cediel, F., Etayo, F. and Cáceres, C., 1997. Distribución de facies sedimentarias y su marco tectónico durante el Fanerozóico en Colombia. IV Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Cartagena, Septemb. 14-17, Tomo II: 34-37. Céspedes S. and Peña L., 1995. Relaciones estratigráficas y ambientes de depósito de las formaciones del Terciario Inferior aflorante entre Tunja y Paz de Rio (Boyacá), Tesis Pregrado Geología, Univ. Nacional de Colombia Depart. Geoc., Santafé de Bogotá, 50 p. Chandra, U., 1981. Focal mechanism solutions and their tectonic implications for the Eastern AlpineHimalayan Region. In H. K. Gupta and F. M. Delany (Eds.), Zagros. Indu Kush. Himalaya. Geodynamic evolution. Geodynamic Series volume 3. Am. Geoph. Union, Washington, D. C. Geol. Soc. Am., Boulder Colorado: 243-271. Chapman, D. S., 1986. Thermal gradients in the continental crust, In: J.B. Dawson, D.A., Carswell, J. Hall, and K.H., Wedepolhl (eds.), The Nature of the Lower Continental Crust . Geological Society Spec. Publ., 24: 63-70. Cheilletz, A., Giuliani, G. and Arhan, T., 1993. Late Eocene-Oligocene episode in the Eastern Cordillera of Colombia viewed by emerald dating. II Symp. Internat. Géodynamique Andine, ISAG 93, Oxford, ORSTOM. Paris, p 473-476. Cheilletz, A., Giuliani, G., Branquet, Y., Laumonier, B., Sánchez, M.A.J., Rueda, F., Féraud, G. and Arhan, T., 1997. Datation K-Ar et 40Ar/39Ar à 65 M.a. des gisements d’émeraude du district Chivor-Macanal, argument en faveur d’une déformation précoce dans la Cordillère Orientale de Colombie. C.R. Acad. Sci., Séries II, Paris, 324(5): 369-377. Chigne, N. and Rojas, E., 1997. El piedemonte de la Cordillera Oriental de Colombia y de los Andes de Mérida: Estilos estructurales y consideraciones sobre la génesis y migración de hidrrocarburos. VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas. Mem. Tomo I: 457-477.
268
References
Choukroune, P. and ECORS Team., 1989. The ECORS Pyrenean deep seismic profile reflection data and the overall structure of an orognic belt. Tectonics 8(1): 23-39. Choukroune, P., Pinet, B., Roure, F. and Cazes, M., 1989. Major Hercynian structure along the ECORS Pyrenees and Biscaye lines. Bull. Soc. Géol. France. Cisternas, A. and Gaulon, R., 1984. Síntesis sismotectónica de nordeste de Venezuela, Revista de Geofísica, Madrid, 40(1): 3-10. Clavijo, J., 1985. La secuencia de la Formación Los Santos en la Quebrada Piedra Azul: registro de una hoya fluvial evanescente. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter IV, Ingeominas Publicación Geológica Especial 16, Bogotá, 18 p. Clavijo, J., 1995. Mapa geológico de Colombia, plancha 75 – Aguachica Escala 1:100.000, Memoria explicativa. Ingeominas, Bucaramanga, 48 p. Clavijo, J., Barbosa, G., Bernal, L.E., and others, 1992. Mapa geológico plancha 75 Aguachica, Scale 1:100.000. Bucaramanga. Cloetingh, S., 1988. Intra-plate stress: a new element in basin analysis. In: K.L. Kleinspehn and C. Paola (eds.), Frontiers in sedimentary geology –New perspectives in basin analysis. SpringerVerlag, New York, NY, p. 205-230. Cloetingh, S. and Banda, E., 1992. Europe’s lithosphere –physical properties: mechanical structure. In: Blundell, D., Freeman, R., and S. Mueller (eds.), A continent revealed – the European geotraverse. Cambridge University Press, European Science Foundation, Cambridge, p. 80-91. Cloetingh, S. and Kooi, H., 1992. Intraplate stress and dynamical aspects of rift basins. In: P. A. Ziegler (ed.), Geodynamics of rifting, v. II. Thematic discussions, Tectonophysics, 215: 167-185. Cloetingh, S. and Burov, E., 1996. Thermomechanical structure of European continental lithosphere; constraints from rheological profiles and EET estimates. Geophys. Jour. Int., 124: 695-723. Cloetingh, S., McQueen, H. and Lambeck, K., 1985. On a tectonic mechanism for regional sea level variations. Earth Plan. Sci. Lett., 75: 157-166. Cloetingh, S., Sassi, W. and Horvath, F., (eds.), 1993. The origin of sedimentary basins: Inference from quantitative modeling and sedimentary basin analysis. Tectonophysics, 226, 518 p. Cloetingh, S., Eldholm, O., Larsen, B. T., Gabrielsen, R. H. and Sassi, W., 1994. Dynamics of extensional basin formation and inversion: introduction. Tectonophysics, 240: 1-9. Cloetingh, S., Burov, E. and Poliakov, A., 1999. Lithosphere folding: Primary response to compression? (from central Asia to Paris Basin). Tectonics, 18(6): 1064-1083. Cloos, M., 1993. Lithospheric buoyancy and collisional orogenesis: subduction of oceanic plateaus, continental margins, island arcs, spreading ridges and seamounts. Geol. Soc. Am. Bull., 105: 715-735. Cobbold, P. R. and Percevault, M. N., 1983. Spatial integration of strains using finite elements. Jour. Struct. Geol., 5: 299-305. Cobbold, P. R., Renzoni, G. and Ulloa, C., 1988. Cenozoic tectonics of the Cordillera Oriental: A working hypothesis. Ingeominas Internal Report. Bogotá, 8 p. Colletta, B., Hébrard F., Letouzey, J., Werner, P. and Rudkiewikz, J. L., 1990. Tectonic style and crustal structure of the Eastern Cordillera (Colombia), from a balanced cross section. In: J. Letouzey (ed.), Petroleum and tectonics in mobile belts. Editions Technip, Paris 11990, p. 81100. Colletta B., Letouzey, J., Faure, J. L. and Casero, P., 1995. Kinematic and physical modeling of thrusts. Comparisons with the Colombian foothills structures. VI Congreso Colombiano del Petróleo. Bogotá, Abs. 1 p. 151-154. Colletta, B., Roure, F., De Toni, B., Loureiro, D. and Passalacqua, H., 1997. Tectonic inheritance, crustal architecture, and contrasting structural styles in the Venezuelan Andes. Tectonics, 16(5): 777-794. Cooper, M. A., Addison, F. T., Alvarez, R., Coral, M., Graham, R. H., Hayward, A. B., Howe, S., Martinez, J., Naar, J., Peñas, R., Pulham, A. J. and Taborda, A., 1995. Basin development and tectonic history of the Llanos Basin, Eastern Cordillera, and Middle Magdalena Valley, Colombia, A.A.P.G. Bull., 79 (10): 1421-1443.
269
References
Coral, C., 1985. Contribución al estudio de la actividad sísmica en Santander (Colombia), Memorias VI Congreso Latinoamericano de Geología, Tomo II: 271-292. Coral, C. and Sarmiento, L., 1986. Resultados preliminares del estudio de parámetros focales en la región de Santander (Colombia), Revista CIAF, Bogotá, 11(1-3): 435-455. Corredor, F., 1997. Evidence for a passive roof duplex along the Northeastern thrust front of Eastern Cordillera, Colombia, and implications for oil exploration. VI Simposio Exploración Petrolera en las Cuencas Subandinas, Cartagena, Tomo II: 42-47. Corrigan, H.T., 1989. Introduction to the geology of the Sabana and southern protions. 26 th Field Conference. Texaco. In: Colom. Soc. Petrol. Geol. Geoph. (Ed.), Geological Field Trips, Colombia 1980-1989. Colombian Society Petroleum Geologists and Geophysicists, Bogotá, 312-332, 1 appendix. Cortes, R. and De la Espriella, R., 1984. Contibución al conocimiento del Paleozóico Superior en la sección Quetame-Villavicencio. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 16(30): 83-101. Covey, M. C. and Dengo, C. A., 1987. Structural evluation of exploration leads in the Paz de Río and Tunja blocks, Eastern Cordillera, Colombia. Ecopetrol, Bogotá, ISN 12535, Informe 2310. Cross, T. A. and Pilger, R. H., 1982. Controls of subduction geometry, location of magmatic arcs, and tectonics of arc and back-arc regions. Geol. Soc. Am. Bull. 93: 545-562. Cross T. A., Anderson, D. S., McDonough K. J., Ramón, J. C., Sanchez, A. and Olaya, I., 1996a. Stratigraphic framework of the Middle Magdalena Basin, Report prepared for Ecopetrol and Instituto Colombiano del Petróleo by Strategic Stratigraphy, Inc Evergreen, Colorado, 39 p. Cross T. A., Anderson, D. S., McDonough K. J., Ramón, J. C., Sanchez, A. and Olaya, I., 1996b. Middle Magdalena Basin stratigraphic play assessment, Phase 2, Report prepared for Ecopetrol and Instituto Colombiano del Petróleo by Strategic Stratigraphy, Inc Evergreen, Colorado, 8 p. Cuervo, E. A. and Ramírez, A., 1985. Estratigrafía y ambiente de sedimentación de la Formación Cacho en los alrededores de Bogotá. Tesis Pregrado Geología. Univ. Nacional de Colombia, Depto. de Geociencias, Bogotá, 100 p. Czaplewski, N. J., 1989. Miocene bats from La Venta fauna, Colombia. In: W. D., Krause and R. D., Kenneth (eds.), Abstracts for papers: Forty-ninth annual meeting, Assoc. of Vertebrate Paleont. Jour. Vertebrate Paleont. 9(3) Supple. p. 18A. Daconte, R. and Salinas, R., 1980. Geología de las planchas 66, Miraflores y 76, Ocaña, Colombia. Ingeominas, Bogotá, Informe 1844: 1-105. Dahlstrom, C. D. A., 1969. Balanced cross-sections. Can. Jour. Earth Sci., 6: 743-757. Daly, M. C., 1989. Correlations between Nazca/Farallon Plate kinematics and forearc basin evolution in Ecuador. Tectonics, 8: 769-790. Daniels, A. T., 1991. Stress and strain in the Northern Andes from earthquake data. M.Sc. Thesis, Univ. of South Carolina, Columbia, S.C., 100 p. Dashwood, M. F. and Abbotts, I. L., 1990. Aspects of the petroleum geology of the Oriente Basin, Ecuador, In: J. Brooks (Ed.), Classic petroleum provinces. Geol. Soc. Spec. Publ. London, 50: 89-117. Davis, J. H. and Blanckenburg, F., 1995. Slab breakoff: A model of lithosphere detachment and its test in the magmatism and deformation of collisional orogens. Earth and Planet. Scien. Lett., 129: 85-102. Davy, Ph. and Gillet, Ph., 1986. The stacking of thrust slices in collision zones and its thermal consequences. Tectonics, 5: 913-929. De Freitas, M. G., Francolin, J. B. L. and Cobbold, P. R., 1997. The structure of the axial zone of the Cordillera Oriental, Colombia. IV Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas.Asociacion Colombiana Geologos y Geofísicos del Petróleo, Bogotá, p. 38-41. De Jong, K., 1991. Tectono-metamorphic studies and radiometric dating in the Betic Cordilleras (SE Spain) – with implications for the dynamics of extension and compression in the Western Mediterranean Area. Ph.D. Thesis, Vrije Universiteit, Amsterdam, 204 p. De la Espriella, R. and Cortes, R., 1982. Contribución a la estratigrafía del Grupo Quetame. IV Congr. Colombiano de Geología. Cali, (In press) 15 p.
270
References
Dengo, C. A. and Covey, M. C., 1993. Structure of the Eastern Cordillera of Colombia: Implications for trap styles and regional tectonics, A.A.P.G. Bull., 77(8): 1315-1337. Deségaulx, P., Roure, F. and Villien, A., 1990. Structural evolution of the Pyrenees: Tectonic heritage and flexural behaviour of the continental crust. In J Letouzey (ed.), Petroleum and tectonics in mobile belts. IFP Exploration and production research conferences, Editions Technip, Paris, p. 31-48. Deségaulx, P., Kooi, H. and Cloetingh, S., 1991. Consequences of foreland basin development on thinned continental lithosphere: Applications to the Aquitane Basin (SW France). Earth Planet. Sci. Lett., 106: 116-132. Dewey, J. F., 1980. Episodicity, sequence, and style at convergent plate boundaries. In: D.W. Strangeway (ed.), The continental crust and its mineral deposits, Spec. Paper Geol. Assoc. Canada, 20: 553-573. Dewey, J. F. and Pindell, J. L., 1985. Neogene block tectonics of eastern Turkey and northern South America; continental applications of the finite difference method. Tectonics, 4(1): 71-83. Dewey, J. W., 1972. Seismicity and tectonics of western Venezuela, Bull. Seism. Soc. Am., 62: 1.7111.752. Diaz L., 1994a. Distribución de las facies siliciclásticas correspondientes a la Formación Arenisca Tierna y equivalents en el Valle Superior del Magdalena. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter IV, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 15 p. Diaz L., 1994b. Reconstrucción de la Cuenca del Valle Superior del Magdalena, a finales del Cretácico. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter XI, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 13 p. Díaz, J. M. and Sotelo, C. I., 1995. Análisis estructural de la Falla de Boyacá en un área al oeste de los Municipios de “Paipa-Duitama”. Tesis Pregrado Geología, Univ. Nacional de Colombia, Depart. Geoc., Bogotá, 75 p. Dickey, P. A., 1941. Pre-Cretaceous sediments in Cordillera Oriental of Colombia. Bull. A.A.P.G., 25(9): 1789-1795. Dickinson, W. R., 1985. Interpreting provenance relations from detrital modes of sandstones. In: G. P., Zuffa (ed.), Provenance of arenites. NATO ASI Series C. Mathematical and Physical Sciences 148: 333-361. Divies, R., 1997. FOLDIS, un modèle cinèmatique de bassins sédimentaires par éléments discrets associants plis, failles, érosion/sédimentation et compaction. Ph.D. Thesis. Univ. of Grenoble I, Grenoble, 220 p. Dorado, J., 1984. Contribución al conocimiento de la estratigrafía de la Formación Brechas de Buenavista (límite Jurásico – Cretácico), región noroeste de Villavicencio (Meta). Geología Colombiana, Bogotá, 17: 7-40. Dueñas, H. and Castro, G., 1981. Asociación palinológica de la Formación Mesa en la región de Falán, Tolima, Colombia. Geología Norandina., Bogotá, 3: 27-36. Duncan R. A. and Hargraves R. B., 1984. Plate tectonic evolution of the Caribbean region in the mantle reference frame. In: W.E. Bonini, R.B. Hargraves and R. Shagam (eds). The Caribbean-South American plate boundary and regional tectonics, Geol. Soc. Am., Mem., 162: 81-93. Duque-Caro, H., 1979. Major estructural elements and evolution of northwestern Colombia. In: J.S. Watkins, L. Montadert, and P.W. Dickerson. (Eds.), Geological and geophysical investigations of continental margins. A.A.P.G. Mem. 29: 329-351. Duque-Caro, H., 1984. Structural style, diapirism, and accretionary episodes of the Sinú-San Jacinto terrane, southwestern Caribbean borderland. Geol. Soc. Am. Mem. 162: 303-316. Duque-Caro, H., 1990. The Chocó block in the northwestern corner of South America: Structural, tectonostratigraphic, and paleogeographic implications. Jour. of South Am. Earth Sci., 3(1): 7184. Duval, B. C. and Valdés, G. E., 1995. Giant oil fields of the ´80s associated with an 'A' subduction in S. America. Oil and Gas Jour., July 1995: 67-70. Ebinger, C. J., 1989. Geometric and kinematic development of border faults and accommodation zones, Kivu-Rusizi rift, Africa. Tectonics, 8: 117-133.
271
References
Ecopetrol., 1970. Mapa Geológico Geológico de los Llanos Orientales (Área Pisba - Támara), Esc.1:50.000. Esc.1:50.000. Ecopetrol-ICP., 1995. Área volcanera: evaluación técnica de subsuelo. Ecopetrol-ICP, Division de Exploración y Explortación, Explortación, EXE, Piedecuesta, 42 p. Ecopetrol-ICP., 1996. Informe Proyecto “Estratigrafía del Terciario del Valle Medio del Magdalena”. Compact disc., Piedecuesta. Ecopetrol, Vicepresidencia de Exploración., 1997. Topes de formaciones de pozos del Valle del Magdalena. Informe interno, Santafé de Bogotá, 100 p, Ecopetrol and Beicip., 1995. Cuenca de los Llanos Orientales, Estudio geológico regional, Santafé de Bogotá, Volume 1, Text, 198 p., Volume 2, figures. Ecopetrol, Esso and Exxon Exploration Company., 1994. Integrated technical evaluation Santander sector Colombia., 1991-1994. 1991-1994. Report, text and figures. Final Report, Houston, 38 p., 49 figs. Ellis, S. and Beaumont, C., 1999. Models of convergent boundary tectonics, implications for the interpretation interpretation of Lithoprobe data, Canadian Jour. of Earth Sci., 36(10): 1.711-1.741. England, P. and Molnar, P., 1990. Surface uplift, uplift of rocks, and exhumation of rocks. Geology, 18: 1173-1177. England, P. C. and Thompson, A. B., 1986. Pressure-temperature-time Pressure-temperature-time paths of regional metamorphism metamorphism I. Heat transfer during the evolution of regions of thickened continental continental crust. Jour. Petrol., 25: 894-928. ESRI - HOCOL., 1989. Proyecto de Evaluación Técnica del Flanco Oriental de los Valles Medio y Alto del Magdalena. Informe Geológico de Ecopetrol Nro. 2424. ESRI and Ecopetrol., 1994. Geology and hydrocarbon potential of the Cordillera Oriental Colombia, 4 volumes. Earth Sciences and Resources Institute, University of South Carolina. ESRI Technical Report 94-07-418, Columbia, SC., 500 p. Estrada, A., 1972. Geology and plate tectonics history of the Colombian Andes. M.Sc. Thesis Stanford Univ., California, 115 p. Etayo-Serna, F., 1968. El sistema Cretáceo en la región de Villa de Leiva y zonas próximas. Geología Colombiana, Bogotá, 5: 5-74. Etayo-Serna, F., 1979. Zonation of the Cretaceous of Central Colombia by ammonites. Publicación Geológica Especial Ingeominas, Bogotá, 2: 1-186. Etayo-Serna, F., 1985a. Documentación paleontológica del Infracretácico de San Félix y Valle Alto, Cordillera Central. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XXV, Ingeominas Publicación Publicación Geológica Especial 16, Bogotá, 7 p. Etayo-Serna, F., 1985b. El límite Jurásico Cretácico en Colombia. In: F. Etayo-Serna and F. LaverdeMontaño (eds.), Proyecto Crétacico, contribuciones. Chapter XXII, Ingeominas Publicación Geológica Especial 16, Bogotá, 4 p. Etayo-Serna, F., 1985c. Paleontología estratigráfica del Sistema Cretácico en la Sierra Nevada del Cocuy. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones . Chapter XXIII, Ingeominas Publicación Geológica Especial 16, Bogotá, 47 p. Etayo-Serna, F., 1994. Epílogo: A modo de historia geológica del Cretácico del Valle Superior del Magdalena. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter XX, Univ. Nacional de Colombia, Ecopetrol, Ecopetrol, Bogotá, 6 p. Etayo-Serna, F. and Laverde-Montaño, F., (eds.), 1985. Proyecto Crétacico, contribuciones. Ingeominas, Publ. Geol. Esp. 16, Bogotá, 200 p. Etayo-Serna, F. and Rodríguez G., 1985. Edad de la Formación Los Santos. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XXVI, Ingeominas Publicación Publicación Geológica Especial 16, Bogotá, 13 p. Etayo-Serna, F. and Florez, J. M., 1994. Estratigrafía y estructura de la Quebrada Calambe y el cerro El Azucar, Olaya Herrera Tolima. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior hapter XII, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 23 p. del Magdalena.C Etayo, F., Renzoni, G. and Barrero, D., 1976. Contornos sucesivos del mar Cretácico en Colombia. Primer Congreso Colombiano de Geología, Mem., Bogotá, p. 217-252. Etayo-Serna, Etayo-Serna, F., Barrero, D., Lozano, H., Espinosa, A., González, H., Orego, A., Zambrano, F., Duque, H., Vargas, R., Núñez, A., Álvarez, J., Ropaín, C., Ballesteros, I., Cardozo, E., Forero, H.,
272
References
Galvis, N., Ramírez, C. and Sarmiento, L., 1983. Mapa de terrenos geológicos de Colombia, Ingeominas Publicación Geológica Especial 14, Bogotá, 235 p. Eurocan., 1988. Informe operacional y técnico para Ecopetrol 1988. Contrato de Asociación Tunja. Ecopetrol Informe No 2133, Bogotá, 20p. Fabre, A., 1981. Estratigrafía de la Sierra Nevada del Cocuy, Boyacá y Arauca, Cordillera Oriental (Colombia). Geología Norandina, Bogotá, 4: 3-12. Fabre, A., 1983a. La subsidencia de la Cuenca del Cocuy (Cordillera Oriental de Colombia) durante el Cretáceo y el Terciario Inferior. Primera parte: Estudio cuantitativo de la subsidencia. Geología Norandina, Bogotá, Bogotá, 8: 49-61. Fabre, A., 1983b. La subsidencia de la Cuenca del Cocuy (Cordillera Oriental de Colombia) durante el Cretáceo y el Terciario Inferior. Segunda parte: Esquema de evolución tectónica. Geología Norandina, Bogotá, Bogotá, 8: 21-27. Fabre A., 1985a. Dinámica de la sedimentación Cretácica en la región de la Sierra Nevada del Cocuy (Cordillera Oriental Oriental de Colombia). In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XIX, Ingeominas Publicación Geológica Especial 16, Bogotá, 20 p. Fabre, A., 1985b. Subsidencia y maduración de la materia orgánica; un modelo de la evolución de la Cordillera Oriental y los Llanos durante el Cretácico y el Terciario (abstract). Mem. IV Congr. Latinoamericano de Geología., Bogotá, 2: 421-422. Fabre, A., 1986. Géologie de la Sierra Nevada del Cocuy (Cordillère Orientale de Colombie). Thèse Université de Genève de Docteur es Sciences de la Terre, No 2217, Genève, 394 p. Fabre, A., 1987. Tectonique et géneration d’hydrocarbures: Un modèle de l’évolution de la Cordillère Orientale de Colombie et du Bassin des Llanos pendant le Crétacé et le Tertiaire. Arch. Sc. Genève, 40 (Fasc. 2): 145-190. Fabre, A. and Delaloye, M., 1983. Intrusiones básicas Cretácicas de la Cordillera Oriental. Geología. Norandina, Bogotá, Bogotá, 6:19-28. Fajardo, A., 1995. 4-D stratigraphic architecture and 3-D reservoir fluid-flow model of the Mirador Formation, Cusiana Field, foothills area of the Cordillera Oriental, Colombia. MSc. Thesis Colorado School of Mines, Golden, 171 p. Fajardo A., Rubiano J. L. and Reyes A., 1993. Estratigrafía de secuencias de las rocas del Cretáceo Tardío al Eoceno Tardío en el sector central de la cuenca de los Llanos Orientales, Departamento Departamento del Casanare, Report ECP-ICP-001-93, Piedecuesta, Santander, 69 p. Fajardo, G. and Chamorro, M. E., 1994. Interpretación geológica y geofísica Proyecto Tunja-94. Ecopetrol, Informe Geol. 3046. Santafé de Bogotá. Fajardo-Peña, G., 1998. Structural analysis and basin inversion evolutionary model of the Arcabuco, Tunja and Sogamoso regions, Eastern Cordillera, Colombia. M.Sc. Thesis, University of Colorado, Boulder, 114 p. g eology. John Wiley, New York, 589 p. Faure, G., 1986. Principles of isotope geology Feininger, T., 1970. The Palestina Fault, Colombia. Geol. Soc. Am. Bull., 81(4): 1201-1216. Feininger, T., 1980. Cretaceous and Paleogene geologic history of coastal Ecuador. Geologishe Rundschau, Bd. 69 (Heft. 3): 849-874. Feininger, T., 1985. Allochthonous terranes in the Andes of Ecuador and northwestern Perú. Geological Survey of Canada, I Observatory Crescent, Ottawa, Ontario. Canada KIA OY3. Feininger, T., 1986. Allochthonous Allochthonous terranes in the Andes of Ecuador and northwestern Perú. Can. Jour. of Earth Sci., 24: 266-278. Findlay, A. L., 1988. Cortes geológicos regionales en los Andes Venezolanos y sus implicaciones petroleras y estructurales. estructurales. Presentación por afiches. III Simposio Bolivariano Bolivariano Exploración Exploración Petrolera en las Cuuencas Subandinas, Caracas, Memorias, Tomo I: 401-429. Flint, S., Turner, P., Jolley, E. J. and Hartley, A. J., 1993. Extensional tectonics in convergent margin basins: an example example from the Salar de Atacama, Chilean Andes. Geol. Soc. Am. Bull. 105: 603617. Flórez J, M. and Carrillo G. A., 1994. Estratigrafía de la sucesión litológica basal del Cretácico del Valle Superior del Magdalena. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena.C hapter II, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 26 p.
273
References
Flüeh, E. R., 1973. Additional gravity interpretation along the active plate boundary in southweast, Colombia. In: J. E., Ramirez and L. T.Aldrich (eds.), La transición oceano-continente en al suroeste de Colombia, Proyecto Cooperativo Internacional Nariño, Univ. Javeriana, Bogotá, p. 199-204. Flüeh, E. R., Milkereit, B., Meissner, R., Meyer, R. P., Ramírez, J. E., Quintero, J. and Udias, A., 1981. Observaciones de refracción sísmica en el Noroeste Colombiano en la latitud 5.5° N. In: Goberna, J.R. (ed.), Investigaciones geofísicas sobre las estructuras oceano-continentales del Occidente Colombiano, Proyectos Cooperativos Internacionales Nariño II 1976, Nariño III 1978. Instituto Geofísico Univ. Javeriana, Bogotá, p. 83-96. Föllmi K. B., Garrison R. E., Ramirez P. C., Zabrano-Ortíz, Kennedy W. J. and Lehner B. L., 1992. Cyclic phosphate-rich successions in the Upper Cretaceous of Colombia. Palaeogeography, Palaeoclimatology, Palaeoecology, 93: 151-182. Forero, H., 1958. Fuentes termales de Colombia. Informe 1295, Servicio Geológico Nacional, Bogotá, 25 p. Forero, H. and Sarmiento L., 1985. La facies evaporítica de la Formación Paja en la región de Villa de Leiva. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XVII, Ingeominas Publicación Geológica Especial 16, Bogotá, 16 p. Freymueller, J. T., Kellogg, J. N. and Vega, V., 1993. Plate motions in the North Andean region. J. Geophys. Res. , 98 (B12): 21.853-21.863. Galindo, V., Garzón, H., Rueda, M. C. and Brown, P., 1994. Evaluación del potencial de entrampamiento estructural en el Cretáceo (Cuenca del Valle Supeior del Magdalena), Ecopetrol, Vicepresidencia de exploración y producción. Gerencia de Exploración , 22 p. Anexo Regional overview of the Upper Magdalena Valley: Tectonic evolution by P. Brown. 26 p. Ecopetrol Informe Geológico, Geológico, Santafé de Bogotá, Bogotá, 15 p. Galvis, N. and Rubiano J., 1985. Redefinición estratigráfica de la Formación Arcabuco, con base en análisis facial. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, Publicación Geológica Especial 16, Bogotá, 16 p. contribuciones. Chapter VII, Ingeominas Publicación García-Castellanos, D., 1998. Desarrollo de modelos numéricos de flexión litosférica. Ph.D. Thesis. Institut de Ciències de la Terre Jaume Almera. Consejo Superior de Investigaciones Científicas (CSIC), Barcelona, 166 p. García-González. M., Velandia-Guevara, J. and Galindo-Torres, G., 1997. El potencial de generación de gas y petróleo de la Formación Guaduas de la Cordillera Oriental, y de la Formación Guachinte-Ferreira de el Valle del Cauca. VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Cartagena, Abs. 14. George. R. P., Pindell, J. L. and Cristancho, J. H., 1997. Eocene paleostructure of Colombia and implications for history of generation and migration of hydrocarbons. IV Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas: Asociación Colombiana Geologos y Geofísicos del Petróleo, Bogotá, Tomo I: 133-139. Geotec., 1976. Mapa geológico de Colombia. Scale 1: 1.000.000. Ed. Geotec, Bogotá, Colombia. Geotec., 1992. Facies distribution and tectonic setting through the Phanerozoic of Colombia. A regional synthesis combining outcrop and subsurface data presented in 17 consecutive rock-time slices. Bogotá.100 p. Geotec., 1994. Stratigraphy and tectonic study of the Upper and Middle Magdalena Basins Colombia, Report for Maxus, Bogotá, 55 p. Germeraad, J. H., Hoping, C. A. and Muller, J., 1968. Palynology of Tertiary sediments from tropical area. Rev. Paleobotan. Palynology, 6: 139-348. Geyer, O., 1976. La fauna de amonitas del perfil típico de Morrocoyal (Liásico inferior). I Congr. Colombiano de Geología, Bogotá, Mem.: 111-134. Geyer, O., 1982. Comparaciones estratigráficas y faciales en el Triásico Norandino. Geología. Norandina, Bogotá, Bogotá, 5: 27-31. Gibbs, A. D., 1987. Linked tectonics of the Northern North Sea basin. In: C. Beaumont and A. J. Tankard (eds.), Sedimentary basins and basin forming mechanisms. Can. Soc. Petrol. Geol., Mem. 12: 163-171.
274
References
Gîrbacea, R. and Frisch, W., 1998. Slab in the wrong place: Lower lithospheric mantle delamination in the last stage of the eastern Carpathian subduction retreat. Geology, 26(7): 611-614. Goetze, C. and Evans, B., 1979. Stress and temperature in the bending lithposhere as constrained by experimental rock mechanics. Geophys. J. Roy. Astr. Soc., 59: 463-478. Goldsmith, R., Marvin, R. F. and Mehnert, H. H., 1971. Radiometric ages in the Santander Massif, Eastern Cordillera, Colombian Andes, U.S. Geol. Surv. Prof. Paper, Reston, V.A. P0750-D: D44-D49. Gomez, E., 1997. Quantified Cenozoic evolution of Cenozoic deformation of the Middle Magdalena (Colombia). A.A.P.G. Bull., 81(10): 1774. Gómez, E., 1999. Geodinámica, tectonoestratigrafía, cronología y evolución termal de cuencas sedimentarias complejas y de sus áreas fuente. Métodos modernos de análisis cuantitativo y aplicaciones, implicaciones para exploración y producción de hidrocarburos. Curso-Taller con salida de campo. Santafé de Bogotá, Bogotá, Nov. 26 –30., 55 p. Gómez E. and Pedraza P., 1994. El Maastrichtiano de la región Honda Guaduas, límite N del Valle Superior del Magdalena: Registro sedimentario de un delta dominado por ríos trenzados. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter III, Univ. Nacional de Colombia, Colombia, Ecopetrol, Ecopetrol, Bogotá, 20 p. Gómez, E. and Jordan, T., 1998. Relative timing of Cenozoic deformation of the Middle Magdalena Valley Basin and bounding mountain ranges, Colombia A.A.P.G./SEPM Annual Meeting, Abstr. with Progr., Salt Lake City, Utah. Gomez, E., Jordan, T., Hegarty, K. and Kelley, S., 1999. Diachronous deformation of the Central and Eastern Andean cordilleras of Colombia and syntectonic sedimentation in the Middle Magdalena Valley Basin. Fourth ISAG, Gottingen, Germany, 4-6 October, p. 287-290. Gradstein, F. M. and Ogg, J. G., 1996, A Phanerozoic time scale. Episodes, Episodes, 9(1-2): 3-5. Gratier, J. P., Guillier, B. and Delorme, A., 1991. Restoration and balance of a folded and faulted surface by best-fitting of finite elements: Principle and applications. Jour. Struct. Geol., 13: 11115. Grupo de Geoquímica-ICP., 1994. Evaluación geoquímica Cuenca Valle Superior del Magdalena. Informe final presentado a la Gerencia de Exploración Ecopetrol. Proyecto Valle Superior del Magdalena, Piedecuesta, 195p. Guatame, R. A. and Lara, H. Y., 1995. Análisis geoquímico de las formaciones pre-Albianas Fómeque y Tibasosa en un área al norte de Tunja entre los Municipios de Pesca, Nobsa, Santa Rosa de Viterbo, Belén y Betéitiva. Tesis Pregrado Geología, Univ. Nacional de Colombia, Depto. Geoc., Bogotá, 100 p. Guerrero, J., 1993. Magnetostratigraphy of the upper part of the Honda Group and Neiva Formation. Miocene uplift of the Colombian Andes. Ph.D. Thesis, Duke University, 108 p. Appendix Orthogonal Vector plots, 108 p. Guerrero, J. and Sarmiento, G., 1996. Estratigrafía Estratigraf ía física, palinologíca, sedimentológica y secuencial del Cretácico Superior y Paleoceno del Piedemonte Llanero, Implicaciones en Exploración Petrolera. Geología Colombiana, Colombiana, 20: 3-66. Guillande, M. R., 1988. Evolution Méso-Cénozoï Méso-Cénozoï que d’une vallée vallée intercordilléraine intercordilléraine andine: La Haute Haute Vallée du Rio Magdalena (Colombie), (Colombie), Ph.D. Thesis Univer. Paris 6, Paris, P aris, 352 p. Ceratost reon boussingaultid Guzmán, G., 1985. Los griféidos infraCretácicos Aetostreon couloni y Ceratostreon e la Formación Rosablanca, como indicadores de occilaciones marinas, In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XII, Ingeominas Publicación Geológica Especial 16, Bogotá, 16 p. Hall, R., Alvarez, J. and Rico, H., 1972. Geología de los departamentos de Antioquia y Caldas (subzona II-A). Boletín Geológico Ingeominas, Bogotá, 20(1): 1-85. Haq, B. U., U. , Hardenbol, J. and Vail, P. R., 1987. Chronology of fluctuating sea levels since the Triassic Triassic (250 Myr ago to Present). Science, 235: 1156-1167. Harry, D. L. and Sawyer, D. S., 1992. Basaltic volcanism, mantle plumes, and the mechanics of rifting: The Paraná flood basalt province of South America. Geology 20: 207-210.
275
References
Hathon, L. and Espejo, I., 1997. Detrital modes of Tertiary units, Central Middle Magdalena Basin, Colombia. IV Simposio Bolivariano, Exploración Petrolera en las Cuencas Subandinas. Cartagena. Cartagena. Abs., 16. Hayashida, A., 1984. Paleomagnetic study of the Miocene continental deposits in La Venta badlands, Colombia. Hébrard F., 1985. Les foothills de la Cordillère Orientale de Colombie entre les rios Casanare et Cusiana. Evolution géodynamique depuis d’Eo Crétacé. Ph.D. Thesis, Univ. Pierre et Marie Curie, Paris, 162 p. Hébrard, F., Leikine, M., Bourgois, J., Ricateau, R. and Charitat, P., 1987a. Metamorphisme et diagenèse des séries Crétacées et Tertiaires du versant est de la Cordillère Orientale de Colombie, un exemple d’évolution thermique liée à l’ouverture d’un bassin au Crétacé. Eclog. Geol. Helv. Hébrard, F., Leikine, M., Bourgois, J., Ricateau, R. and Charitat, P., 1987b. Métamorphisme et diagenèse des series Crétacées et Tertiaires du versant est de la Cordillère Orientale de Colombie, un exemple d’évolution thermique liée à l’ouverture d’un bassin au Crétacé. II Simposio Bolivariano Bolivariano Exploración Petrolera en las Cuencas Subandinas, Bogotá, Abs. Helmens, K. F., 1988. Late Pleistocene glacial sequence in the area of the high plain of Bogotá (Eastern Cordillera, Colombia). Palaeogeography, Palaeoclimatology, Palaeoecology, 67: 263-283. Helmens, K. F., 1990. Neogene-Quaternary geology of the high plain of Bogotá, Eastern Cordillera, Colombia (stratigraphy, paleoenvironments and landscape evolution). Diss. Botanicae 163, J. Cramer, Berlin, 202 p. Helmens, K., Barendregt, R., Randolph, E., Baker, J. and Andriessen, P. A. M., 1997. Magnetic polarity and fission-track chronology chronology of a late Pliocene-Pleistocen Pliocene-Pleistocenee paleoclimatic paleoclimatic proxy record in the tropical Andes. Quaternary Research, 48: 15-28. Henneberg, H. E., 1983. Geodetic control of neotectonics neotectonics in Venezuela. Tectonophysics, 97: 1-15. Hill, R. I., 1993. Mantle plumes and continental tectonics. Lithos 30: 193-206. Hirsfeld S. E. and Marshall, L. G., 1976. Revised fauna list of the La Venta Fauna (Friasian, Miocene) of Colombia, South America. Jour. Paleontology, 50: 433-436. Hoogiemstra, H., 1984. Vegetational and climatic history of the high plain of Bogotá, Colombia: a continuous record of the last 3.5 million years. Diss. Botanicae, 79, 368 p. Hooghiemstra, H., 1989. Quaternary and Upper-Pliocene glaciations and forest development in the tropical Andes: evidence from long high-resolution pollen record from the sedimentary basin of Bogotá, Colombia. Palaeogeography, Palaeoclimatology, Palaeoecology, 72: 11-26. Hoorn, C., 1988. Quebarada Mochuelo, type locality of the Bogotá Formation: A Sedimentological, petrographical etrographical and palynological palynological study. study. Hugo de Vries Lab., Univ. of Amsterdam, Amsterdam, 21 p. Hossack, Hossack, J., Martínez, J., Estrada, C. and Herbert, R., 1999. Structural evolution of the Llanos fold and thrust belt, Colombia. In: K. McClay (ed.), Thrust Tectonics 99 Meeting, Royal Halloway Univ. of London, April 26-29, June 1999, Program, London, p. 110. Howe, M. W., 1969. Geologic studies of the Mesa Group (Pliocene?), Upper Magdalena Valley, Colombia. Ph.D. Thesis, Princeton Univ., Princeton, 67 p. Howe, M. W., 1974. Nonmarine Neiva Formation (Pliocene?), Upper Magdlena Valley, Colombia: regional tectonism. tectonism. Geol. Soc. Am., Bull., 85: 228-246. Hubach, E., 1957. Estratigrafía de la Sabana de Bogotá y alrededores. Boletín Geológico, Servicio Geológico Nacional, Nacional, Bogotá, 5(2): 93-112. ICP-GEX-DYA, Grupo Neiva, 1995. Estudio hidrodinámico de la Formación Caballos en la Subcuenca de Neiva. Cuenca Valle Superior del Magdalena. Informe final, Piedecuesta, 39 p. Irving, E. M., 1971. La evolución estructural de los Andes mas septentrionales de Colombia. Boletín Geológico Ingeominas, Ingeominas, Bogotá, 19 (2): 1-190. Isacks, B. and Molnar, P., 1971. Distribution of stress in the descending lithosphere from a global survey of focal mechanism solutions of mantle earthquakes, Rev. Geoph., 9: 103-174. Jaillard , E. P., Solar, P., Carlier, G. and Mourier, T., 1990. Geodynamic evolution of the northern and central Andes during early to midle Mesozoic times: a Tethyan model. Jour. Geol. Soc., London, 147: 1009-1022.
276
References
Janssen, M., Stephenson, R. A. and Cloetingh, S., 1995. Temporal and spatial correlations between changes in plate motions and the evolution of rifted basins in Africa. Geol. Soc. Am. Bull., 107: 1317-1332. Jaramillo, J. M., 1978. Determinación de las edades de algunas rocas de la Cordillera Central de Colombia por el método de huellas de fisión. 2do. Congreso Colomb. Geol. Resúmenes, Bogotá, P. 19-20. Jaramillo, J. M., 1981. Determinación de las edades de algunas rocas de la Cordillera Central de Colombia por el método de huellas de fisión. Boletín Ciencias de la Tierra, Medellín, (5-6): 145-146. Jaramillo L, Roa. E. and Torres M., 1993. Relaciones estratigráficas entre las unidades arenosas del Paleogeno (Paleoceno) del piedemonte llanero y la parte media de la Cordillera Oriental. Tesis Pregrado Geología Univ.Nacional de Colombia, Bogotá, 80 p. Jaramillo C. and Yépez O., 1994. Palinostratigrafía del Grupo Olini (Coniaciano-Campaniano), Valle Superior del Magdalena, Colombia. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena.C hapter XIII, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 20 p. Jarvis, G. T. and McKenzie, D., 1978. Sedimentary basin formation with finite extension rates. Earth Planet. Sci. Lett., 48: 42-52. Jones, P. B., 1995. Geodynamic evolution of the Eastern Andes, Colombia –An alternative hypothesis. In: A.J. Tankard, R. Suárez S., and H. J., Welsnik (eds.), Petroleum basins ba sins of o f South America: A.A.P.G. Mem., 62: 647-658. Jordan, T. and Alonso, R., 1987. Cenozoic stratigraphy and basin tectonics of the Andes Mountains, 20° - 28 °s outh latitude. A.A.P.G. Bull., 71: 49-64. Jordan, T. and Gomez, E., 1996. Proposal of Middle Magdalena Valley stratigraphic study of Tertiary sediments to Ecopetrol. Introduction, Introduction, Ecopetrol, Bogotá, 12 p. Julivert, M., 1958a. La morfoestructura de la zona de mesas al SW de Bucaramanga (Colombia S.A.). Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 1: 7-43. Julivert, M., 1958b. Geología de la zona tabular entre San Gil y Chiquinquirá, Cordillera Oriental, Colombia. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 2: 33-45. Julivert, M., 1959. Geología de la vertiente W del macizo de Santander en el sector de Bucaramanga. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 3: 15-33. Julivert, M., 1960. Geología de la región occidental de García Rovira. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 5: 5-32. Julivert, M., 1961. Las estructuras del Valle Medio del Magdalena y su significación. Univ. Industrial de Santander, Bucaramanga, Boletín Geológico, 6: 33-52. Julivert, M., 1962. La estratigrafía de la Formación Guadalupe y las estructuras por gravedad en la serranía de Chía, Sabana de Bogotá. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 11: 5-21. Julivert, M., 1968. Colombie. Lexique stratigraphique International. Precambrian, Paleozoique et Mesozoique. Paris, vol. V, Fasc. 4a, premiere partie: 651 p. Julivert, M., 1970. Cover and basement tectonics in the Cordillera Oriental of Colombia, and a comparison comparison with other folded chains. Geol. Soc. Am. Bull., 81: 3623-3646. Julivert, M., Barrero, D. and Navas, J., 1964. Geología de la Messa de Los Santos. Boletín Geológico Uiv. Industrial de Santander, Bucaramanga, 18: 5-11. Kaila, K. L., 1981. Structure and seismotectonics of the Himalaya-Pamir Hindukush region and the Indian Plate boundary. In H. K. Gupta and F. M. Delany (Eds.), Zagros. Indu Kush. Himalaya. Geodynamic evolution. Geodynamic Series volume 3. Am. Geoph. Union, Washington, D. C. Geol. Soc. Am., Boulder Colorado: Colorado: 272-293. Kammer, A., 1993a. Las Fallas de Romeral y su relación con la tectónica de la Cordillera Central. Geología Colombiana, Bogotá, 18: 27-46. Kammer, A., 1993b. Steeply dipping basement faults and associated structures of the Santander Massif, Eastern Cordillera, Colombian Andes. Geología Colombiana, Bogotá, 18: 47-64. Kammer, A., 1996. Estructuras y deformaciones del borde oriental del Macizo de Floresta. Geología Colombiana, Colombiana, Bogotá, 21: 65-80.
277
References
Kammer, A., 1999a. Structural styles of the folded Bogotá segment, eastern Cordillera Oriental of Colombia. Fourth ISAG, Gottingen, Germany, Oct. 4-6, 1999, Extended Abstracts, p. 381-384. Kammer, A., 1999b. Observaciones acerca de un origen traanspresivo de la Cordillera Oriental. Geología Colombiana, Bogotá, 24: 29-53. Kammer, A., 2001. Tectónica extensiva Jurásica al margen continental Norandina y el origen de la Falla de Bucaramanga. Resumen Coloquio Triásico-Jurásico en Colombia. Ingeominas, Univ. Indust. Santander, Sociedad Colombiana de Geología. Bucaramanga, 1 p. Kammer, A. and Mojica, J., 1995. Los pliegues de la Barrera de Guataqui–Girardot: Producto de un despegue de la Cobertera Cretácica de la Cordillera Oriental?. Geología Colombiana, Bogotá, 19: 33-48. Kammer, A. and Mojica, J., 1996. Una comparación de la tectónica de basamento de las cordilleras Central y Oriental. Geología Colombiana, Bogotá, 20: 93-106. Karner, G. D. and Watts, A. B., 1983. Gravity anomalies and flexure of the lithosphere at mountain ranges. J. Geophys. Res. , 88: 10449-10477. Kay R. F., Madden, R. H. and Guerrero, J., 1989. Nuevos hallazgos de monos in el Mioceno de Colombia. Ameghiana, 25(3): 203-212. Keal, J. E., 1985. Source rock studies of the Eastern Cordillera adjacent to the Llanos Basin, Colombia. Mem. II Simposio. Bolivariano Exploración Petrolera en las Cuencas Subandinas, Bogotá, 2: 119. Keeley, M. L. and Súarez, A. E., 1997. Exploring the sub-thrust play along the eastern margin of the Middle Magdalena Vallley: The example of Chawina (Apulo), Cundinamarca, Central Colombia. VI Simposio Bolivariano Exporación en las Cuencas Subandinas, Cartagena de Indias, Mem. Tomo I: 167-184. Kellogg, J., 1984. Cenozoic tectonic history of the Sierra de Perijá, Venezuela-Colombia, and adjacent basins. In : W. E. Bonini, R. B. Hargraves and R. Shagam (eds.), The Caribbean South American plate boundary and regional tectonics, Geol. Soc. Am., Mem., 162: 239-261. Kellogg, J. and Duque, H. S., 1994. Crustal scale deformation in the Eastern Cordillera, Colombia. V Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas. Puerto La Cruz, Memoria: 365-372. Kellogg, J., Ogujiofor, I. J. and Kansakar, D. R., 1985. Cenozoic tectonics of the Panama and North Andes blocks, Mem. Sixth Latin American Geological Congress., Bogotá, 1: 40-59. Kellogg, J., Godley, V. M., Ropaín, C., Bermúdez, A. and Aiken, C. L. V., 1991. Gravity field of Colombia, eastern Panamá and adjacent marine areas. Geol. Soc. Am. Map and Chart Series, MCH070. Kellogg, J., Ojeda, G., Duque, H., Roth, S., Scott, M., Zhao, Ji, K., Diaz, D. and Gladdish, D., 1995. Colombia geophysical project Eastern Cordillera Technical report. An interdisciplinary project to interpret the geology and geopysics of Colmbia's Eastern Cordillera, foothills and Vichada. Andean Geophysical Laboratory Univ South Carolina, Columbia S.C., 166 p. Kerr, A. C., Mariner, G. F., Arndt, N. T., Tarney, J., Nivia, A., Saunders, A. D., Storey, M. and Duncan, R. A., 1996. The petrogenesis of Gorgona komatiites, picrites and basalts: new field, petrographic and geochemical constraints. Lithos 37: 245-260. Kerr, A. C., Mariner, G. F., Tarney, J., Nivia, A., Saunders, A. D., Thirlwall, M. F. and Sinton, C. W., 1997. Elemental and isotopic constraints on the petrogenesis and origin of Cretaceous basaltic terranes in Western Colombia. Jour. Petrol. 38:677-702 Khalturin, V. I., Rautian, T. G. and Molnar, P., 1977. The spectral content of Pamir-Hindu Kush intermediate depth earthquakes; evidence for a high-Q zone in the upper mantle, J. Geophys. Res. , 82(20): 2.931-2.943. Kirby, S. H., 1983. Rheology of the lithosphere. Rev. Geophys. Space Phys., 21: 1458-1487. Kluth, Ch., F., Laad, R., De Armas, M., Gómez, L. and Tilander, N., 1997. Different structural styles and histories of the Colombian foreland: Castilla and Chichimene oil fields areas, east-central Colombia. VI Simpolsio Bolivariano Exploracion en las Cuencas Subandinas, Cartagena de Indias, Mem. Tomo II: 185-197. Kohn, B. P., Shagam, R., Banks, P. O. and Burkley, E. A., 1984. Mesozoic-Pleistocene fission track ages on rocks of the Venezuelan Andes and their tectonic implications. In: W.E. Bonini, R.B.,
278
References
Hargraves and R. Shagam (eds.), The Caribbean-South American plate boundary and regional tectonics. Geol. Soc. Am. Mem., 162: 365-384. Kooi, H., 1991. Tectonic modelling of extensional basins: the role of lithosphere flexure, intraplate stresses and relative sea-level change. Ph.D. Thesis, Vrije Universiteit, Amsterdam. 183 p. Kooi, H., Cloetingh, S. and Burrus, J., 1992. Lithospheric necking and regional isostasy at extensional basins: part 1, Subsidence and gravity modeling with an application to the Gulf of Lions margin (SE France). J. Geophys. Res. , 97: 17553-17571. Kovas, E. J., Rodgers, D. A. and Binger, J. H., 1982. Seismic interpretation of back thrusts and displacement transfer zone between in échelon thrust faults, Middle Magdalena Valley, Colombia. V Congreso Latino-Americano de Geología, Argentina, Actas 1: 565-582. Kuhry, P., Helmens, K. and Rutter, N. W., 1994. Magnitude and duration of four stadial intervals between 22 and 8 Ka. In the tropical Andes of Colombia (South America). Geol. Soc. Am. 1994 Annual Meeting Abstracts with programs. Geol. Soc. Am. 26(7), p. 446. Kusznir, N. J. and Karner, G., 1985. Dependence of the flexural rigidity of the continental lithosphere on rheology and temperature. Nature, 316: 138-142. Kusznir, N. J. and Ziegler, P. A., 1992. Mechanics of continental extension and sedimentary basin formation: a simple-shear/pure-shear flexural cantilever model. In: P.A. Ziegler (ed.), Geodynamics of rifting, v. III, Thematic discussions, Tectonophysics, 215: 117-131. Kusznir, N. J., Marsden, G. and Egan, S. S., 1991. A flexural cantilever simple-shear/pure-shear model of continental lithosphere extension: application to the Jeanne d’Arc Basin, Grand Banks and Viking Graben, North Sea. In: A. M. Roberts, G., Yielding and B. Freeman (eds.), The geometry of normal faults . Geol. Soc., London, Spec. Publ. 56: 41-60. Lance-Le Cornec S., 1998. 1D and 2D themal modeling of sandstone reservoirs. Incidence of the circulation of meteoric and hydrothermal water. In: Toro, J., Bordas-LeFloch, N., Le CornecLance, S., Roure, F., Bénard, F., Lafargue, E., Sassi, W., Robion, P., Aubourg, C., Frizon de La Motte, D., Guilhaumou, N., Report on the Colombian transects Subtrap Consortium (Subthrust reservoir appraisal), Paris, Chapter 2, 20p. Laubscher, H. P., 1987. The kinematic puzzle of the Neogene Northern Andes. In J. P Schaer and J. Rodgers, (eds.), The anatomy of mountain ranges, Chapter 11, Princeton University Press, Princeton, New Jersey, p. 211-227. Laubscher, H., 1990. The regional tectonics of the thrustbelts bordering the Llanos and the Middle and Upper Magdalena Valley, Colombia. Report to Hocol. Geological Institute of the University Basel, Switzerland, 35 p. Laverde, F., 1979. Espesor estratigrafía y facies de la Formación Guaduas en algunos sitios del Cuadrángulo K-11. Tesis Pregrado Geología, Univ. Nacional de Colombia, Bogotá, 92 p. Laverde F., 1985. La Formación Los Santos: un depósito continental anterior al ingreso marino del Cretácico. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter XX, Ingeominas Publicación Geológica Especial 16, Bogotá, 24 p. Laverde, F., 1989. Stratigraphy of the Tertiary sequence southwest of Bogotá, Colombia, northeastern Upper Magdalena Valley – western border of the Cordillera Oriental. M.Sc. Thesis, Univ. of South Carolina, Columbia, SC., 66 p. Laverde F. and Clavijo, J., 1985. Análisis facial de la Formación Los Santos, según el corte Yo y Tu (Zapatoca). In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter VI, Ingeominas Publicación Geológica Especial 16, Bogotá, 9 p. Laverde, F. and Ramírez, C., 1987. Unpublished stratigraphic column, 1:5000 scale, Area 4. In: R. Ressetar, and C.E. Macellari). ESRI Tech. Rept. 88-0006, 40 p. Leeds, A., 1977. Velocidades del manto en la región Colombia-Ecuador. In: S. J. Ramírez and L.T. Aldrich (eds.), La Trancisión océano-continente en el suroeste de Colombia. Proyecto Cooperativo Internacional Nariño. Inst. Geof. Univ. Javeriana, Bogotá, p. 237-238. Linares, R., 1996. Structural styles and kinematics of the Medina area, Eastern Cordillera, Colombia, M.Sc. Thesis, Univ. of Colorado, 104 p. Lyon-Caen, H. Armijo, R. and Cifuentes, I., 1990. Recent deformation of the Ecuadorian sub-Andes (abstract), Eos Trans, AGU, 71, p 1560.
279
References
Macia, C., 1995. El Magmatismo Tríasico Superior – Jurásico en Colombia. Breve síntesis, contribution to IGCP 322 Jurassic events in South America, Geología Colombiana, Bogotá, 20: 163-164. Macía, C. and Mojica, J., 1981. Nuevos puntos de vista sobre el magmatismo Tríasico Superior (Fm Saldaña), Valle Superior del Magdalena. Colombia. Zbl. Geol. Palaeont. Stuttgart, 1(3-4): 243251. Macía, C., Mojica, J. and Colmenares, F., 1985. Consideraciones sobre la importancia de la paleogeografía y las áreas de aporte precretácicas en la prospección de hidrocarburos en el Valle Superior del Magdalena, Colombia. Geología Colombiana, Bogotá, 14: 49-70. Malavielle, J., 1984. Modélisation expérimentale des chevauchements imbriquées: Application aux chaînes de montagnes. Bull. Soc. Géol. France, 7, XXVI(1): 129-138. Mandl, G., 1988. Mechanics of tectonic faulting ; models and basic concepts, Developments in Structural Geology 1. Elsevier, Amsterdam, 407 p. Mann P. and Burke, K., 1984. Neotectonics of the Caribbean. Rev. Geophys., 22: 309-362. Mann P., Schubert, C. and Burke, K., 1990. Review of Caribbean neotectonics. In G. Dengo and J Case (eds.), The Caribbean Region, vol. H, The geology of North America. Geol. Soc. Am. Boulder, p. 307-338. Mantilla, M., 2000, Presentación de resultados interpretación área Lisama. In: Modelamiento estructural de los campos Cretáceos del VMM. Unpublished report Instituto Colombiano del Petroleo, Piedecuesta, Santander. 12 p. Marksteiner, R. and Aleman, A. M., 1997, Petroleum systems along the fold belt associated to the Marañón – Oriente Putumayo (MOP) foreland basins. VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas. P. 63-74. Martínez, J. I., 1990. Estratigrafía de la plancha 227. Ingeominas Informe Interno, Bogotá, 120 p. Martínez W., Rubio, Méndez, A. and Taborda, A., 1998. Ideas sobre la Geología estructural del Noreste de Colombia y su relación con los campos de petróleo. III Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas Caracas, p. 505-522. Maughan, E. K., Zambrano, F. O., Mojica, G., Abozagalo, M., Pachón, F. P. and Duran, R. R., 1979. Paleontologic and stratigraphic relations of phosphate beds in Upper Cretaceous rocks of the Cordillera Oriental, Colombia. U.S. Geol. Surv. Open File Report 79-1525, 97 p. Mayorga M. and Vargas, M., 1995. Caracterización geoquímica y facial de las rocas potencialmente generadoras de hidrocarburos del Cretáceo y Terciario inferior, Cordillera Oriental y Piedemonte. Tesis Pregrado Geología, Univ. Nacional de Colombia, Bogotá, 150 p. Maze, W. B., 1984. Jurassic la Quinta Formation in the Sierra de Perijá, northwestern Venezuela: geology and tectonic environment of red beds and volcanic rocks. In: W. E. Bonini, R. B., Hargraves and R. Shagam (eds.), The Caribbean-South American plate boundary and regional tectonics. Geol. Soc. Am. Mem., 162: 263-282. McCollough,C. N. and Carver, J. A., 1992. The giant Caño Limón field, Llanos Basin, Colombia. In: M.T. Halbouty (Ed.), Giant oil and gas fields of the decade 1978-1988 Conference. A.A.P.G. Mem. 54: 175-195. McCourt, W. J., Feininger, T. and Brook, M., 1984. New geological and geochronological data from the Colombian Andes: continental growth by multiple accretion. Jour. Geol. Soc. London, 141: 831-845. McKenzie, D., 1978. Some remarks on the development of sedimentary basins. Earth Planet. Sci. Lett., 40: 25-32. McKenzie, D. and Bickle, M. J., 1988. The volume and composition of melts generated by extension of the lithosphere. Jour. Petrol., 29: 625-679. McLaughlin Jr, D. H., 1972. Evaporite deposits of Bogotá Area, Cordillera Oriental, Colombia. A.A.P.G. Bull., 56(11): 2240-2259. McLaughlin, D. H. and Arce Jr., M., 1969. Mapa geológico del cuadrángulo K-11 Zipaquirá, Scale 1:100.000, Ingeominas, Bogotá. Mégard, F., 1987. Cordilleran Andes and marginal Andes: A review of Andean geology north of the Arica elbow (18º S). In: J.W.H. Monger, J. Francheteau (Eds), Circum-Pacific orogenic belts
280
References and evolution of the Pacific Ocean Basin.
International Lithosphere Program Contribution, Geodynamic Series, Washignton, D.C., V. 18: 71-95. Meier, B., Schwander, M. and Laubscher, H. P., 1987. The tectonics of Táchira: a sample of North Andean Tectonics. In J. P. Schaer and J. Rodgers, (eds.), The anatomy of mountain ranges, Chapter 12, Princeton University Press, Princeton, New Jersey, p. 229-237. Meissner, R. O., Flüeh, E. R., Stibane, F. and Berg, E., 1973. Dynamics of the active plate boundary in southwest Colombia according to recent geophysical measurements. In: J. E., Ramírez and L. T. Aldrich (eds.), La transición océano-continente en el suroeste de Colombia, Proyecto Cooperativo Internacional Nariño, Univ. Javeriana, Bogotá, p. 157-168. Meissner, R. O., Flüh, E. R., Stibane, F. and Berg, E., 1976. Dynamics of the active plate boundary in southwest Colombia according to recent geophysical measurements. Tectonophysics 35: 115136. Mejía, R. M. V. and Giraldo, R. J., 1993. Estudio petrográfico de la Formación Picacho en un área al Noreste de Sogamoso, Boyacá. Tesis pregrado Geología, Univ. Nacional de Colombia, Bogotá, 80 p. Mendivelso, L. D., 1982. Aspectos fotogeológicos y estratigráficos del Cretáceo en la regón de Itaibe, Valle Superior del Magdalena. Tesis Pregrado geología. Univ. Nacional de Colombia, Depto. de Geociencias, Bogotá, 116 p. Mendoza H., 1985. La Formación Cumbre, modelo de transgresión marina rítmica, de comienzos del Cretácico. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter IX, Ingeominas Publicación Geológica Especial 16, Bogotá, 17 p. Meschede, M. and Frisch, W., 1998. A plate tectonic model the Mesozoic and Early Cenozoic history of the Caribbean Plate. Tectonophysics, 296(3-4): 269-291. Meyer, R. P., Mooney, W. D., Hales, A. L., Helsley, C. E., Woollard, G. P., Hussong, D. M., Kroenke, L. W. and Ramírez, J. E., 1973. Refraction observation across a leading edge, Malpelo Island to the Colombian Cordillera Occidental. In: J.E., Ramirez and L. T.Aldrich (eds.), La transición océano-continente en al suroeste de Colombia, Proyecto Cooperativo Internacional Nariño, Univ. Javeriana, Bogotá, p. 83-136. Miall, A. D., 1992. Exxon global chart; an event for every occasion?. Geol. Soc. Am. Bull., 20(9): 787790. Miall, A. D., 1993. Exxon global chart; an event for every occasion?: Reply. Geol. Soc. Am. Bull., 21(3): 284-285. Miller, T. A. and Etayo-Serna, F., 1979. The geology of the Eastern Cordillera, between AguazulSogamoso-Villa de Leiva, 1972, In Geotec (Ed.) Geological Field Trips, Colombia, 1958-1978. Colombian Society Petroleum Geologists and Geophysicists, Bogotá: 349-396. Mojica, J., 1982. Observaciones acerca del estado actual del conocimiento de la Formación Payandé (Triásico superior), Valle superior del Río Magdalena, Colombia. Geología Colombiana, Bogotá, 11: 67-91. Mojica, J., 1983. Direcciones de esfuerzos preteritos y recientes en Colombia. 10a Conferencia Geológica del Caribe, 14-22 Aug, 22 p. Mojica, J., 1985. Estilolitos horizontales en las sedimentitas del Cretácico de la Cordillera Oriental, y su significado tectónico. Geología Colombiana, Bogotá, 14: 7-24. Mojica, J. and Llinas, R., 1984. Observaciones recientes sobre las características del basamento económico del Valle superior del Magdalena en la región de Prado-Rovira (Tolima, Colombia) y en especial sobre la estratigrafía y petrografía del Miembro Chicalá (=parte baja de la Fm. Saldaña). Geología Colmbiana, Bogotá, 13: 81-128. Mojica, J. and Macia, C., 1981. Características estratigráficas y edad de la Formación Yaví, Mesozoico de la región entre Prado y Dolores, Tolima, Colombia. Geología Colombiana, Bogotá, 12: 7-32. Mojica, J. and Sheidegger, A. E., 1981. Diaclasas recientes en Colombia y su significado tectónico. Geología Colombiana, Bogotá, 12: 57-90. Mojica, J. and Scheidegger, A. E., 1984. Fósiles deformados y otras estructuras microtectónicas en la Formación Hiló (Albiano). Alrededores de Sasaima, Cundinamarca (Colombia). Geología Colombiana, Bogotá, 13: 41-54.
281
References
Mojica, J. and Villarroel, C., 1984. Contribución al conocimiento de las uniddes Paleozóicas del área de Floresta (Cordillera Oriental Colombiana, Departamento de Boyacá) y en especial de la Formación Cuche. Geología Colombiana, Bogota, 13: 55-78. Mojica, J. and Franco, R., 1990. Estructura y evolución tectónica del Valle Medio y Superior del Magdalena, Colombia. Geología Colombiana, 17: 41-64. Mojica, J. and Macía, C., 1992. Geological Reconnaissence of the Northern Neiva Basin, Upper Magdalena Valley, Colombia. XXVII Field Conf. 1980, Geol. Field Trips 1980-1989, Colombian Society Petroleum Geologists and Geophysicists. Ediciones Geotec, Bogotá, p. 91120. Mojica, J. and Kammer, A., 1995. Resumen de Jurásico en Colombia, Contribution IGCP 322 Jurassic Events in South America, Geología Colombiana, Bogotá, 20: 160-162. Mojica, J., Macia, C. and Colmenares, F., 1985. Consideraciones sobre la importancia de la paleogeografía y las áreas de aporte precretácicas en la prospección de hidrocarburos en el Valle Superior del Magdalena, Colombia . Geología Colombiana, Bogotá, 14: 49-70. Mojica J., Kammer A. and Ujueta, G., 1996. El Jurásico del sector noroccidental de Suramerica y guía de la excursión al Valle Superior del Magdalena (Nov. 1-4/95), Regiones de Payandé y Prado, Departamento del Tolima, Colombia. Geología Colombiana, Bogotá, 21: Mooney, W. D., Meyer, R. P., Laurence, J. P., Meyer, H. and Ramirez, J. E., 1979. Seismic refraction studies of the Western Cordillera, Colombia. Bull. Seism. Soc. Am. 69: 1745-1761. Mora, A. and Kammer, A., 1999. Comparación de los estilos estructurales en la sección entre Bogotá y los Farallones de Medina, Cordillera Oriental de Colombia. Geol.- Colombiana,Bogotá, 24: 5582. Mora, C., 1996. Evaluación geoquímica. In: Torres M.P., Fajardo, G., Mora, C.A., Bedregal, R.P., Estudio de evaluación conjunta estratigráfica, geoquímica y estructural Bloques Laguna (Occidental), Lanceros (Braspetro) y Tunja (Ecopetrol). Cordillera Oriental.Santafé de Bogotá, 100 p. Mora, C., 1997. Pulsos de generación y expulsión de hidrocarburos a partir de rocas de edad Cretácica en el Valle Medio del Magdalena (Colombia). VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Cartagena, Abs., 22. Mora, C., 2000. Evaluación del potencial de los sistemas petrolíferos en las Cuencas Cretácicas con producción comercial en Colombia (Putumayo, Valle Superior del Magdalena, Valle Medio del lMagdalena, Catatumbo y Llanos Orientales). Maestría en Sistemas Petrolíferos Univ. Federal de Río de Janeiro and Ceenpes/Petrobras (Coppe and Cegeq). Río de Janeiro, 177 p., 4 appendixes. Mora, C. and Posada, C., 2000. Elementos y procesos de los sistemas petrolíferos en las cuencas Cretáceas Colombianas con producción comercial (Putumayo, Valle Superior del Magdalena, Valle Medio del Magdalena, Catatumbo y Llanos Orientales). In prep. Mora, C., Cordoba, F., Luna O., Sarmiento L. F., Bartels H., Rangel A., Giraldo B. N., Reyes J. P. and Magoon L., 1996. Definición de los Sistemas Petrolíferos del Valle Medio del Magdalena. Empresa Colombiana de Petróleos Vicepresidencia de Exploración y Producción, Gerencia de Exploración. Instituto Colombiano del Petróleo. Internal Report, Bogotá, 100 p. Mora, C., Torres, M. P. and Escobar, J., 1997. Potencial generador de hidrocarburos de la Formación Chipaque y su relación estratigráfica secuencial en la zona axial de la Cordillera Oriental (Colombia). VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Cartagena, Abs., 23, Tomo I: 217-237. Morales, L. G., and the Colombian Petroleum Industry., 1956. General geology and oil occurrences of the Middle Magdalena Valley, Colombia. In: L. G. Weeks (ed.), Habitat of the Middle and Upper Magdalena basins, Colombia. Oil - a symposium, A.A.P.G. p. 641-695. Moreno, J. M., 1990a. Stratigraphy of the Lower Cretaceous Rosablanca Formation, west flank, Eastern Cordillera, Colombia. Geología Colombiana, Bogotá, 17: 65-86. Moreno, J. M., 1990b. Stratigraphy of the Lower Cretaceous units central part Eastern Cordillera, Colombia, 13th International Sedimentological Congress, Nothingham, Aug 22-31, 1990. Abstract.
282
References
Moreno, J. M., 1991. Provenance of the Lower Cretaceous sedimentary sequences, central part, Eastern Cordillera, Colombia. Revista Academia Colombiana Ciencias Exactas Físicas y Naturales, 18(69): 159-173. Moreno, J. M. and Concha, A.E., 1993. Nuevas manifestaciones ígneas básicas en el flanco occidental de la Cordillera Oriental, Colombia. Geología Colombiana, 18: 143-150. Moreno, M. and Rubiano, J., 1987, Unpublished stratigraphic column, 1:5000 scale, Area 2. In: R. Ressetar, and C.E. Macellari). ESRI Tech. Rept. 88-0006, 40 p. Moreno J. and Velázquez, M., 1993. Estratigrafía y tectónica en los alrededores del municipio de Nunchía, Departamento del Casanare, Colombia. Tesis Pregrado Geología Univ. Nacional de Colombia, Bogotá, 60 p. Morley, C. K., Nelson, R. A., Patton, T. L. and Munn, S. G., 1990. Transfer zones in the East African rift system and their relevance to hydrocarbon exploration in rifts. A.A.P.G. Bull., 74: 12341253. Mueller, S. and Phillips, R. J., 1991. On the initiation of subduction. J. Geophys. Res. , 96: 651-666. Naar, J. and Coral, M. G., 1993. Modelo tectónico de un área del Piedemonte Llanero al noroeste de Yopal-Casanare (Colombia). Tesis Pregrado Univ. Nacional de Colombia, Bogotá, 33p. Namson, J., Cunningham, R. and Wodcock, G., 1994. Structural geology and hydrocarbon potential of the northern part of the Upper Magdalena Basin, Colombia. V Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Puerto La Cruz, Memoria: 356-364. Nelson, R. A., Patton, T. L. and Morley, C. K., 1992. Rift-segment interaction and its relation to hydrocarbon exploration in continental rift systems. A.A.P.G. Bull., 76: 1153-1160. Nivia, A., 1987. The geochemistry and origin of the Amáime and volcanic sequences, SW Colombia. M. Phil. Thesis, Univ. of Leicester, UK, 164 p. Notestein, F. B., Hubman, C. W. and Bowler, J. W., 1944. Geology of the Barco Concession, Republic of Colombia. Geol. Soc. Am. Bull., 55: 73-1218. Núñez, A., 1986. Petrogénesis del Batolito de Ibagué. Geología Colombiana, Bogotá, 15: 35-45. Nuttal, C. P., 1990. A review of the Tertiary non-marine molluscan faunas of the Pebasian and other inland basins of northwestern South America. Bull. Bri. Mus. Nat. Hist. Geol. 45(2): 165-371. Occidental, Braspetro and Ecopetrol., 1996. Geochemistry and sequence strtigraphy of the source rocks in the Colombia eastern Cordillera. Field trip. Asociación Colombiana Geólogos y Geofísicos del Petróleo, Santafé de Bogotá, 25 p. Occidental, Ecopetrol and Braspetro., 1996. Estudio de evaluación conjunta estratigráfica, geoquímica y estructural bloques Laguna (Occidental), Lanceros (Braspetro) y Tunja (Ecopetrol), Cordillera Oriental. Bogotá, 100 p. Ocola, L., Aldrich, L. T., Gettrust, J. F., Meyer, R. P. and Ramírez, J. E., 1973. Project Nariño I: Crustal structure under southern Colombian-northern Ecuador Andes from seismic refraction data. In: J. E., Ramirez and L. T.Aldrich (eds.), La transición océano-continente en al suroeste de Colombia, Proyecto Cooperativo Internacional Nariño I, Univ. Javeriana, Bogotá, p. 47-70. Ocola, L. C., Aldrich, L. T., Getrust, T. F., Meyer, R. P. and Ramírez, J. E., 1975. Proyecto Nariño I: Crustal structure under southern Colombia-northern Ecuador Andes from seismic refraction data. Bull. Seism. Soc. Am., 65: 1.685-1.695. Ochoa, L. E. and Ponguta, S. Y., 1991. Determinación de esfuerzos tectónicos actuales mediante diversos métodos, especialmente alongación de pozos (“breakouts”) en el Valle Superior del Magdalena, Colombia, Mem. IV Simposio Exploración Petrolera en las Cuencas Subandinas, Bogotá, Trabajo 13. Ojeda, G. and Peña H., 1994. Interpretación de la geología del subsuelo en el área Ortega Chaparral Coyaima (Departamento del Tolima). In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena.C hapter VIII, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 20 p. Okaya, N., Cloetingh, S. and Mueller, St., 1996. A lithosphere cross-section through the Swiss Alps – II. Constraints on the mechanical structure of a continent-continent collision zone. Geoph. Jour. Int. 127, 399-414. Olaya, I. D.and Serrano M.P., 1998, Análisis sísmico estratigráfico del área de Zarzalito en la Cuenca VMM. Informe final. Instituto Colombiano del Petróleo División de Exploración y Producción,
283
References
Area de Estratigrafía, Ecopetrol Vicepresidencia de Adjunta dce Exploración, Gerencia de Prospección de Hidrocarburos, Piedecuesta, 52 p., 6 appendixes. Compact disc. Ord, A. and Hobbs, B. E., 1989. The strength of the continental crust, detachment zones, and the development of plastic instabilities. Tectonophysics, 158: 269-289. Page, W. D. and James, M. E., 1981. The antiquity of the erosion surfaces and late Cenozoic deposits near Medellín, Colombia: Implications to tectonics and erosion rates. Mem. 1º Sem. Cuat. Col., Revista CIAF, 6: 421-454. Pardo Casas, F. and Molnar, P., 1987. Relative motion of the Nazca (Farallon) and South American plates since Late Cretaceous time. Tectonics, 6: 233-248. Parker, R. L., 1972. The rapid calculation of potential anomalies. Geophys. Jour. Roy. Astr. Soc., 31: 447-455. Parkinson, N. and Summerhayes, C., 1985. Synchronous global sequence boundaries. A.A.P.G. Bull., 69, 685-687. Peel, F. J. and Hossack, J. R., 1987. A structural study of the T.E.A. And adjacent areas in the Eastern Cordillera, Colombia. Ecopetrol ISN 9737, Informe 2219 (In 11297), 35 p. Pennington, W. D., 1981. Subduction of the eastern Panamá Basin and seismo-tectonics of northwestern South America, J. Geophys. Res. , 86: 10753-10770. Pennington, W. D., Mooney, W. D., Van Hissenhoven, R., Meyer, H. J., Ramírez, J. E. and Meyer, R. P., 1979. Results of a reconnaissance microearthquake survey of Bucaramanga, Colombia. Geoph. Res. Lett, 6: 65-68. Pérez, O.J., Jaimes, M.A. and Garciacaro, E., 1997, Microseismicity evidence for subduction of the Caribbean plate beneath the South American plate in northwestern Venezuela. J. Geophys. Res. , 102(B8): 17.875-17882. Perez, G. and Salazar, A., 1978. Estratigrafía y facies del Grupo Guadalupe. Geología Colombiana. Bogotá, 10: 7-85. Pilger, R. H., Jr., 1984. Cenozoic plate kinematics subduction and magmatism: South American Andes. Jour. Geol. Soc., London, 141: 793-802. Pimpirev C. T., Patarroyo, P. and Sarmiento, G., 1992. Stratigraphy and facies analysis of the Caqueza Group, a sequence of Lower Cretaceous turbidites in the Cordillera Oriental of the Colombian Andes. Geología Colombiana, Bogotá, 17: 297-308. Pindell J. L., 1993. Regional synopsys of Gulf of Mexico and Caribbean evolution, In Pindell, J. L., and Perkins, B. F., (eds.), GCSSEPM Foundation 13 th Annual Research Conference Proceedings, Mesozoic and early Cenozoic development of the Gulf of Mexico and Caribbean region. SEPM Gulf Coast Section, p. 251-274. Pindell, J. and Dewey, J., 1982. Permo-Triassic reconstruction of western Pangaea and the evolution of the Gulf of Mexico-Caribbean region. Tectonics 1: 179-211. Pindell, J. L. and Barret, S. F., 1990. Geological evolution of the Caribbean region: A plate tectonic perspective. In: G. Dengo and J Case (eds.), The Caribbean region, vol. H, The geology of North America. Geol. Soc. Am. Boulder, p. 405-432. Pindell, J. and Erikson, J., 1993. The Mesozoic margin of northern South America. In: J. Salfity (ed.), Cretaceous tectonics of the Andes, Vieweg Germany, p. 1-60. Pindell J. L. and Tabbutt K. D., 1995. Mesozoic-Cenozoic Andean paleogeography and regional controls on hydrocarbon systems. In A. J., Tankard, R. Suarez and H. J. Welsink (eds.), Petreoleum basins of South America : A.A.P.G. Mem., 62: 101-128. Pindell, J., George, Jr., R. P., Cristancho, J. and Higgs, R., 1997. Clarification of the Late CretaceousPaleogene evolution of Colombia. IV Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Bogotá, Tomo I: 129-132. Polanía, J. H. and Rodríguez, O. G., 1978. Posibles turbiditas del Cretáceo Inferior (Miembro Socotá) en el área de Anapoima (Cundinamarca); una investigación sedimentológica basada en registros gráficos. Geología Colombiana, Bogotá, 10: 87-91. Porta, J., 1965. La estratigrafía del Cretàcico Superior y Terciario en el extremo S del Valle Medio del Magdalena. Univ. Industrial de Santander, Bucaramanga, Boletín Geológico (19): 5-30. Porta, J., 1966. La geología del extremo sur del Valle Medio del Magdalena. Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 22-23: 1-347.
284
References
Posamentier, H, W., Allen, G. P., James, D. P. and Tesson, M., 1992. Forced regressions in a sequence stratigraphic framework: concepts, examples, and exploration significance. A.A.P.G. Bull., 76: 1687-1709. Price, R. A., 1973. Large-scale gravitational flow of supracrustal rocks, southern Canadian Rockies. In: K. de Jong and R. Scholten (eds.), Gravity and tectonics, Wiley, New York. Proceedings of the 7thI nternational conference on basement tectonics, p. 491-502. Puerto, E. Y., 1982. Gradientes geotermicos en las cuencas de los Valles Inferior, Medio y Sueprior del Magdalena y de los Llanos Orientales de Colombia. Tesis Ingeniería de Petróleos Fundasción Universidad de America, Bogotá, 154 p. Pulham, A. J., 1994. The Cusiana field, Llanos Basin, eastern Colombia; high resolution sequence stratigraphy applied to late Paleocene-early Oligocene estuarine, coastal plain and alluvial clastic reservoirs. In: S. D. Johnson (ed.), High resolution sequence stratigraphy; innovation and applications. Univ. of Liverpool, Abstract volume, p. 63-68. Pulham, A. J., 1995. The sedimentology and stratigraphy of the early Tertiary Cusiana field, Colombia, Congreso Colombiano del Petróleo, Bogotá, p. 107-113 Pulido, O., 1979a. Geología de las planchas 135 Sa Gil y 151 Charalá. Departamento de Santander, Ingeominas, Boletín Geológico Ingeominas, Bogotá, 23(2): 42-78. Pulido, O., 1979b. Mapa geológico preliminar de la plancha 135, San Gil. Scale 1:100.000, Ingeominas, Bogotá. Pulido, O., 1980. Mapa geológico preliminar de la plancha 151, Charalá. Scale 1:100.000. Ingeominas, Bogotá. Raasvelt, H. C., 1956. Mapa geológico de la Plancha L-9 Girardot, Map scale 1:200.000, Servicio Geológico Nacional, Bogotá. Raasvelt, H. C., et al., 1957. Mapa geológico de la Plancha M-8 Ataco, Map scale 1:200.000, Servicio Geológico Nacional, Bogotá. Ramirez, J. E. and Aldrich, L. T., (eds.), 1973. La transición océano-continente en al suroeste de Colombia, Proyecto Cooperativo Internacional Nariño, Univ. Javeriana, Bogotá, 313 p. Ramírez, N. and Ramírez, H., 1994. Estratigrafía y origen de los carbonatos del Cretácico Superior en el Valle Superior del Magdalena, Departamento del Huila, Colombia. In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter V, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 15 p. Ranalli, G., 1987. Rheology of the Earth. Allen and Unwin, 366 p. Boston (2 nde dition: 1994). Ranalli, G., 1995. Rheology of the Earth. Chapman & Hall, London, 413 p. Ranalli, G. and Murphy, D.C., 1987. Rheological stratification of the lihtosphere. Tectonophysics, 132: 281-295. Rathke, W. W. and Coral, M. G., 1997. Cupiagua field, Colombia: Interpretation case history of a large, complex thrust belt gas condensate field. VI Simposio Bolivariano Expoloración Petrolera en las Cuencas Subandinas, Cartagena, Abs., 9.Tomo I: 119-128. Renzoni, G., 1965a. Apuntes acerca de la litología y tectónica de la zona al este y sureste de Bogotá. Boletín Geológico, Servicio Geológico Nacional Bogotá, X(1-3):59-79. Renzoni, G., 1965b. Geología del cuadrángulo L-11 Villavicencio, mapa a escala 1:200.000. Servicio Geológico Nacional e Inventario Minero Bogotá. Renzoni, G., 1967. Geología del Cuadrángulo J-12, Tunja. Boletín Geológico, Ingeominas, Bogotá, 24(2): 32-48. Renzoni, G., 1968. Geología del Macizo de Quetame. Geología Colombiana, Bogotá, 5: 75-127. Renzoni, G., 1985a. Paleoambientes de la Formación Tambor en la Quebrada Pujamanes. In: F. EtayoSerna, F. Laverde-Montano (eds.), Proyecto Crétacico, contribuciones. Chapter III, Ingeominas Publicación Geológica Especial 16, Bogotá, 18 p. Renzoni, G.,1985b. La secuencia facial de la Formación Los Santos por la Quebrada Piedra Azul: Registro de una hoya fluvial evanescente. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Cretácico, contribuciones. Chapter IV, Ingeominas Publicación Geológica Especial 16, Bogotá, 18 p.
285
References
Renzoni, G., 1985c. Paleoambientes de la Formación Arcabuco y Cumbre de la Cordillera de Los Cobardes. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter X, Ingeominas Publicación Geológica Especial 16, Bogotá, 14 p. Renzoni, G., 1992. Mapa Geológico de la Plancha 193 de Yopal: Ingeominas, Esc 1:100.000, Bogotá. Renzoni, G., Rosas, H. and Etayo, F., 1967. Mapa geológico, plancha 191, Tunja, Map scale 1:100.000. Ingeominas, Bogotá. Ressetar, R. and Schamel S., 1986. The geology and hydrocarbon potential of the Upper and Middle Magdalena Valleys, Colombia. Ecopetrol, Informe Geol. 1545, Bogotá. Restrepo, J.J., Toussaint, J. F., Gonzalez, H., Gordani, U., Kawashita, K., Linares, E. and Parica, C., 1991. Precisiones geocronológicas sobre el occidente Colombiano. Simposio sobre el magmatismo Andino y su marco tectónico. Programa Internacional de Correlación Geológica UNESCO, Union Internacional de Ciencias Geológicas. Manizales, Colombia, Julio 3-5 de 1991. Manizales, p. 1-21. Restrepo-Pace, P. A., 1989. Restauración de la sección geeológica Cáqueza – Puente Quetame: Moderna interpretación estructural de la deformación del flanco este de la Cordillera Oriental. Tesis Geología (Laureada), Univ. Nal. de Colombia, Santafé de Bogotá, 100 p. Restrepo-Pace, P. A., 1995. Late Precambrian to Early Mesozoic tectoic evolution of the Colombian Andes, based on new geochronological, geochemical and isotopic data. Ph.D. Thesis. Univ. Arizona, 195 p. Restrepo-Pace, P. and Villamil, T., 1998. Assesment of structural styles and estimation of stratigraphic thickness, stretching and crustal shortening in the Eastern Andean Cordillera, Colombia. VI Simposio Exploración Petrolera Cuencas Subandinas, Cartagena, Tomo II: 90-94. Restrepo-Pace, P. A., Colmenares, F., Higuera, C. and Mayorga, M., 1999a. Fold and thrust belt along the western flank of the Eastern Cordillera of Colombia: style, kinematics and timing contraints derived from seismic data and detailed surface mapping. In: K. McClay (ed.), A.A.P.G. Spec. Publ. In press., 25 p. Restrepo-Pace, P. A., Colmenares, F., Higuera, C., Mayorga, M. and Leal, J., 1999b. Fold and thrust belt along the western flank of the Eastern Cordillera of Colombia: style, kinematics and timing contraints derived from seismic data and detailed surface mapping. A.A.P.G. Bull., 82(10), p. 1956. Reyes, J. P., 1994. Estratigrafía de secuencias y potenciales trampas estratigráficas en el área del Meta Llanos Orientales de Colombia. V Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Puerto La Cruz, Memoria: 122-146. Reyes, I. and Reyes V. M., 1976. Geología del yacimiento y variabilidad de las características geoquímicas del mineral de hierro en la región de Paz Vieja (Municipio de Paz de Río, Departamento de Boyacá). Primer Congreso Colombiano de Geología, Memoria, p. 267-324. Rivadeneira, M. V., 1996. Cretaceous-Paleogene stratigraphic sequences and the early Andean orogenic events in the Ecuadorian Oriente Basin. Third ISAG. St-Malo, France, 17-19, Sept./ 1996, Extended abstracts, p. 469-472. Robertson Research Inc., 1983. The northern Llanos of Colombia, hydrocarbon potential and stratigraphical control. Propietary report. Houston, 250 p. Robertson Research Inc., 1986. Llanos Basin, Colombia, a sedimentololgical and stratigraphical correlation study. 290 p. Vol. 1, Text; Vols. 2-4, Appendices, Vols. 5-8, Enclosures. Robertson Research Inc., 1989. Tectonic and stratigraphical study of the Middle and Upper Magdalena Valley, Colombia, 4 Vols. Rodríguez, C. and Rojas, R., 1985. Estratigrafía y tectónica de la serie InfraCretácica en los alrededores de San Félix, Cordillera Central de Colombia. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Cretácico, contribuciones. Chapter XXI, Ingeominas Publicación Geológica Especial 16, Bogotá, 21 p. Rodríguez, E. and Ulloa, C., 1994a. Mapa Geológico de la Plancha 169 de Puerto Boyacá: Ingeominas, Esc 1:100.000, Bogotá. Rodríguez, E. and Ulloa, C., 1994b. Mapa Geológico de la Plancha 189 de La Palma: Ingeominas, Esc 1:100.000 Bogotá.
286
References
Roeder, D. and Chamberlain, R. L., 1995. Eastern Cordillera of Colombia: Jurassic-Neogene crustal evolution. In: A. J. Tankard, R. Suárez S., and Welsnik H. J. (eds.), Petroleum basins of South America: A.A.P.G. Mem., 62: 633-645. Rohrman, M., 1995. Thermal evolution of the Fennoscandian region from fission track thermochronology; an integrated approach. Ph.D. Thesis, Vrije Universiteit, Amsterdam, 168 p. Rolón, L. F. and Carrero, M. M., 1995. Análisis estratigráfico de la sección Cretácica aflorante al oriente del anticlinal de Los Cobardes entre los Municipios de Guadalupe-Chima-Contratación, Departamento de Santander. Tesis pregrado Geología, Univ. Nacional de Colombia, Bogotá, 80 p. Romero, F. H. and Rincón, M. A., 1990. Características petrográficas y geoquímicas de las rocas volcánicas de Iza (Departamento de Boyacá, Colombia). Geología Colombiana, Bogotá, 17: 159-168. Romero, F. H., Schultz, R. A. and Kawashita, K., 2000. Geoquímica del rubidio-estroncio y edad de las esmeraldas Colombianas. Geología Colombiana, Santafe de Bogotá, 25: 221-239. Rosendahl, B. R., 1987. Architecture of continental rifts with special reference to East Africa. Ann. Rev. Earth Planet. Sci. 15: 445-503. Ross M. I. and Scotese C. R., 1988. A hierarchical tectonic model of the Gulf of Mexico and Caribbean region. Tectonophysics, 155: 139-168. Rouby, D., 1995. Restauration en carte des domaines faillés en extension: Méthode et applications. Mem. Geosci. Rennes, Rennes, France, 58, 230 p. Rouby, D., Fossen, H. and Cobbold, P. R., 1996. Extension, displacement, and block rotation in the Larger Gullfaks Area, northern North Sea, determined from map view restoration. A.A.P.G. Bull., 80: 875-890. Rouby, D., Xiao, H. and Suppe, J., 1998. 3D-restoration of complexly folded and faulted surfaces using multiple unfolding mechanisms. Unpublished Report, 20 p. Roure, F., 1998. Subduction of the lihtosphere and crustal balance in orogenic belts. In: S. Cloetingh, G. Ranalli and C.A. Ricci (eds.), Sedimentary basins –models and constraints. Proceedings of the International School Earth and Planetary Sciences, Siena, p. 175-192. Roure, F., Choukroune, P., Berastegui, X., Muñoz, J. A., Matheron, P., Bareyt, M., Séguret, M., Camara, P. and Deramond, J., 1989. ECORS deep seismic data and balanced cross-sections; geometric constraints on the evolution of the Pyrenees, Tectonics, 8(1): 41-50. Roure, F., Colletta, B., De Toni, B., Loureiro, D., Passalacqua, H. and Gou, Y., 1997. Within-plate deformations in the Maracaibo and East Zulia basins, western Venezuela. Marine and Petroleum Geology, 14(2): 139-163. Royden, L. H., 1993a. The tectonic expression of slab pull at continental convergent boundaries. Tectonics, 12: 303-325. Royden, L. H., 1993b. Evolution of retreating subduction boundaries formed during continental collision. Tectonics, 12: 629-683. Royden, L. and Keen, C. E., 1980. Rifting process and thermal evolution of the continental margin of eastern Canada determined from subsidence curves. Earth Planet. Sci. Lett. 51, 343-361. Royden, L. and Karner, G. D., 1984. Flexure of the continental lithosphere beneath Apennine and Carpathian foredeep basins: evidence for an insufficient topographic load. A.A.P.G. Bull., 68: 704-712. Rubiano, J. L., 1989. Petrography and stratigraphy of the Villeta Group, Codillera Oriental, Colombia, South America. M.Sc. Thesis, Univ. South Carolina, Columbia, SC., 96 p. Rubiano, J. L.,1998. Dipmeter paleocurrent determination in the Provincia field, Middle Magdalena Valley. Ciencia Tecnología y Futuro, Piedecuesta, Santander, 1(4): 33-48. Sacks, P. E. and Secor, D. T., 1990. Delamination in collisional orogens, Geology 18: 999-1002. Salazar, A., 1987. Unpublished stratigraphic column, 1:5000 scale, Area 5. In: R. Ressetar, and C.E. Macellari). ESRI Tech. Rept. 88-0006, 40 p. Salazar, A., 1992. Depositional and paleotectonic settings of the Cretaceous sequence, Upper Magdalena Valley, Colombia. PhD. Dissertation, University of South Carolina, Columbia S.C., 200 p.
287
References
Salvador, M., 1991. Gravity field, topography and crustal structure of the North Andes. M.Sc. Thesis Univ. of South Carolina, Columbia SC., 100 p. Santo, T., 1969. Characteristics of seismicity in South America, Bull. Earthquake Res. Inst., Univ. Tokyo, 47: 635-672. Sarmiento, G., 1992. Estratigrafía y medios de depósito de la Formación Guaduas. Ingeominas Bogotá, Boletín Geológico, 32(1): 3-44. Sarmiento, G., 1993. Estratigrafía, palinología y paleoecología de la Formación Guaduas (Maastrichtiano Paleoceno Colombia), Academisch Proefrschrift. Ph.D. Thesis Univ. of Amsterdasm, Amsterdam, 192 p. Sarmiento L. F., 1989. Stratigraphy of the Cordillera Oriental west of Bogotá, Colombia. M.Sc. Thesis University of South Carolina, Columbia, SC.,102 p. Sarmiento, L. F., Cardozo, E., Forero, H. and Ramírez, C., 1985. Importancia de la estratigrafía en la evaluación de las anomalías geoquímicas: caso del área del río Zumbe, Utica (Cundinamarca). In: F. Etayo-Serna and F. Laverde-Montaño (Eds.) Proyecto Cretáceo, contribuciones. Publ. Geol. Esp. Ingeominas, Chapter XXVIII Aplicación económica de los estudios estratigráficos Part C, Bogotá, 10 p. Sarmiento, L. F., Van Wess, J.D. and Cloetingh, S., in prep. Mesozoic and Cenozoic rheological evolution of the lithosphere of the Eastern Cordillera, Colombian Andes. Inferences from tectonic models., 24 p. Sassi, W., Rudkewickz, J. L. and Divies, R., 1998. New methods for integrated modelling of deformation and petroleum generation in fold and thrust belts. A.A.P.G. Annual Meeting, Salt Lake City. Abstracts. Sawyer, D. S., 1985, Brittle failure in the upper mantle during extension of continental lithosphere. J. Geophys. Res. , 90 (B4), 3021-3025. Sawyer, D. S. and Harry, D. L., 1991. Dynamical modeling of divergent margin formation; application to the U. S. Atlantic margin. Mar. Geol. 102: 29-42. Sclater, J. G. and Christie, P. A. F., 1980. Continental stretching; an explanation of the post midCretaceous subsidence of the central North Sea Basin. J. Geophys. Res. , 85: 3711-3739. Schamel, S., 1991. Middle and Upper Magdalena basins, Colombia. In: K.T. Biddle (ed.), Active margins. A.A.P.G. Mem., 52: 283-301. Schelling D., 1994. Stuctural Geology. In: Earth Sciences and Resources Institute (ESRI), 1997, Technical Report 86-0014, Univ. of South Carolina, Columbia, SC. v. 4, 100 p. Schneider, J. F., Pennington, W. D. and Meyer, R. P., 1987. Microseismicity and focal mechanisms of the intermediate-depth Bucaramanga nest, Colombia. J. Geophys. Res. , 92: 13.915-13.926. Schubert, C., 1982. Neotectonics of the Boconó Fault, western Venezuela. Tectonophysics, 52: 205220. Schwabe, E., Toro, G., Kairuz, Ch. and Ferreira, P., 2001. Edades por trazas de fisión de circones provenientes de la Formación Saldaña, Valle Superior del Magdalena. Resumen. Coloquio Triásico-Jurásico en Colombia. Ingeominas, Univ. Indust. Santander, Sociedad Colombiana de Geología. Bucaramanga, Febrero 22-23 de 2001. Bucaramanga, 1 p. Servicio Geológico Nacional, (Compilator from Ecopetrol and other Petroleum Companies)., 1967a. Geología del cuadrángulo H-11, Barrancabermeja, Map scale 1:200.000. Servicio Geológico Nacional, Bogotá. Servicio Geológico Nacional, (Compilator from Ecopetrol and other Petroleum Companies)., 1967b. Geología del cuadrángulo I-11, Cimitarra, Map scale 1:200.000. Servicio Geológico Nacional, Bogotá. Setoguchi, T. and Rosenberg, A. L., 1985. Some new ceboid primates from the La Venta, Miocene of Colombia. Mem., VI Congreso Latinoamericano de Geología, Bogotá, v. 1: 287-298. Setoguchi, T. and Rosenberg, A. L., 1987. A fossil owl monkey from La Venta, Colombia. Nature, 326: 692-694. Shagam, R., 1975. The Northern termination of the Andes. In: A.E. M., Nairn, and H. Stehli (eds.), The ocean basins and margins, v. 3, The Gulf of Mexico and the Caribbean. Plenum Press, New York and London, Chapter 9, p. 325-420.
288
References
Shagam, R., Kohn, B. P., Banks, P. O. Dash, L. E., Vargas, R., Rodriguez, G. I. and Pimentel, N., 1984. Tectonic implications of Cretaceous-Pliocene fission track ages from rocks of the circum-Maracaibo Basin region, western Venezuela and Eastern Colombia. In: W. E. Bonini, R. B. Hargraves and R. Shagam (eds.), The Caribbean South American plate boundary and regional tectonics, Geol. Soc. Am., Mem. 162, p. 385-412. Shemenda, A. I., 1993. Subduction of the lithosphere and back-arc dynamics: Insigths from physical modeling. J. Geophys. Res. , 98: 16.167-16.185. Sibson, R. H., 1995. Selective fault reactivation during basin inversion: potential for fluid redistribution through fault-valve action. In: J. G. Buchanan and P. G. Buchanan (eds.), Basin inversion. Geol. Soc. London, Spec. Publ., 85: 3-19. Sinton, C. W., Duncan, R. A., Storey, M., Lewis, J. and Estrada, J. J., 1998. An oceanic flood basalt province within the Caribbean plate. Earth and Planet. Sci. Lett. 155: 221-235. Sleep, N. H., 1971. Thermal effects of the formation of Atlantic continental margins by continental break-up. Geophys. J. Roy. Astr. Soc., 24: 325-350. Smith, G. A. and Landis, C. A., 1995. Intra-arc Basins. In: C. J. Busby and R. V. Ingersoll (eds.), Tectonics of sedimentary basins, Blackwell Science, Cambridge, Massachusetts, Chapter 7, p. 263-298. Steckler, M. S. and Watts, A. B., 1978. Subsidence of the Atlantic-type continental margin off New York. Earth Planet. Sci. Lett., 41:1-13. Steinmann, M., Seward, D. and Hungerbuehler, D., 1996. Thermotectonic history of the Andes, South Ecuador: evidence from fission-track dating. Third ISAG. St-Malo, France, 17-19 Sept./ 1996, Extended abstracts, p. 501-503. Stephenson, R. A. and Cloetingh, S., 1991. Some examples and mechanical aspects of continental lihtospheric folding. Tectonophysics, 188: 27-37. Steuer, M. R., Bentham, P. A. and Latimer, G. H., 1997. Structural style and emplacement of the Opón anticline, Middle Magdalena Valley, Colombia. VI Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas. Cartagena, Mem. Tomo II: 257-262. Streckeisen, A., 1976. To each plutonic rock its proper name. Earth Sci. Rev., 12(1): 1-33. Suárez, G., Molnar, P. and Burchfiel, B. C., 1983. Seismicity, fault planes solutions, depth of faulting, and active tectonics of the Andes of Perú, Ecuador, and southern Colombia, J. Geophys. Res. , 88(B12): 10.403-10428. Suárez M., 1982. Estratigrafía y Facies de la Formación “Arenisca de El Limbo” en el Borde Oeste de la Cuenca de los Llanos Orientales. Tesis Pregrado. Universidad Nacional de Colombia. Bogotá. Pag. 74. Suárez, M., 1997. Tectonoestratigrafía del Terciario y potencial petrolífero de la Formación La Paz, Cuenca del Valle Medio del Magdalena (parte norte), Colombia. VI Simposio Exploración Petrolera en las Cuencas Subandinas, Tomo II: 95-98. Suppe, J.,1983. Geometry and kinematics of fault-bend folding. Am. Jour. Sci. 238: 684-721. Suppe, J. and Medwedeff, D. A., 1990, Geometry and kinematics of fault-propagation folding. In: P. Jordan, T., Noack, S., Schmid and D. Bernoulli (eds.), The Hans Laubsher volume. Eclog. Geol. Helv., 83(3): 409-454. Taboada, A., Rivera, L.A., Fuenzalida, A., Cisternas, A., Philip, H., Bijwaard, H., Olaya, J. and Rivera, C., 2000, Geodynamics of the northern Andes: Subductions and intracontinental deformation (Colombia). Tectonics, 19 (5): 787-813. Taboada, A., Rivera, L. A., Fuenzalida, A., Cisternas, A., Philip, H., Castro, J. E. and Rivera, C., 1999. Geodynamics of the northern Andes: Intra-continental subduction and the Bucaramanga seismicity nest (Colombia). Fouth ISAG, Goettingen, Germany, 4-6 Oct/1999, p. 719-724. Takemura, K. and Danhara, T., 1983. Fission-track age of pumices included in the Gigante Formation, north of Neiva, Colombia. Overseas Research Reports of New World Monkeys, Kioto Univ, Japan, v. 3: 13-15. Tamaki, K. and Honza, E., 1991. Global tectonics and the formation of marginal basins: Role of the western Pacific. Episodes, 14: 224-230.
289
References
Tellez, N. and Navas, J., 1962. Interferencia de las direcciones de los pliegues cretácico-terciarios entre Coello y Gualanday (Valle Superrior del Magdalena). Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 9: 45-61. Ter Voorde, M., 1996. Tectonic modelling of lithospheric extension along faults, implications for thermal and mechanical structure and basin stratigraphy. Ph.D. Thesis, Vrije Universiteit, Amsterdam, 197 p. Testamarck, J. S., Perdomo, J. L. and Serrano, I., 1994. Estilos estructurales del flanco suroriental de la Sierra de Perijá, Venezuela. V Simposio Exploración Petrolera en las Cuencas Subandinas, Puerto La Cruz, Memoria: 314-316. Thouret, J. C., 1988. La Cordillère Centrale des Andes de Colombie et ses bordures: Morphogenèse Plio-Quaternaire et dynamique actuelle et récente d’une cordillère volcanique englacée. Ph.D. Thesis, Univ. Joseph Fourier, Grenoble, France. Thouret, J. C., Murcia, A., Salinas, R. and Vatin Pérignon, N., 1985. Cronoestratigrafía mediante dataciones K-Ar y 14C de los volcanes compuestos del complejo Ruíz-Tolima y aspectos volcano-estructurales del Nevado del Ruíz (Cordillera Central, Colombia). VI Congreso Latinoamericano de Geología, Bogotá, 1: 385-454. Toro, G., Popeau, G., Hermelin, M. and Schwabe, E., 1999. Chronology of the volcanic activity and regional thermal events: A contribution from the tephrochronology in the north of the Central Cordillera, Colombia. Fourth ISAG. Geottingen, Germany, 4-6 Oct./ 1999, Extended Abstracts, p. 761-763. Toro, J., 1990. The termination of the Bucaramanga Fault in the Cordillera Oriental, Colombia. M.Sc. Thesis Univ. of Arizona, Arizona, 60 p. Toro, J., 1992. Interpretación estructural del área Gualanday, Valle Superior del Magdalena. Ecopetrol Vicepresidencia de Exploración y Producción, Gerencia de Exploración. Proyecto Gualanday No 20, Ecopetrol, Informe Geológico, Santafé de Bogotá, 40 p. Toro, J., 1998. Geodynamic evolution of the Eastern Cordillera. In: Toro, J., Bordas-LeFloch, N., Le Cornec-Lance, S., Roure, F., Bénard, F., Lafargue, E., Sassi, W., Robion, P., Aubourg, C., Frizon de La Motte, D., Guilhaumou, N., Report on the Colmbian transects Subtrap Consortium (Subthrust reservoir appraisal), Paris, Chapter 1, 12p. . Torres, C., 1992. In situ. stress in the North Andes from borehole breakout analysis. M.Sc. Thesis University of South Carolina, Columbia S.C., 153 p. Toussaint, J. F., 1978. Grandes rasgos geológicos de la parte septentrional del noroccidente Colombiano. Boletín Ciencias de la Tierra. Univ. Nal. Medellín, 1: 1-147. Toussaint, J. F., 1993. Evolución geológica de Colombia, 1, Precámbrico, Paleozóico. Univ. Nacional de Colombia, Medellín, 229 p. Toussaint, J. F., 1995a. Hipótesis sobre el marco geodinámico de Colombia durante el Mesozóico temprano, Contribution to IGCP 322 Jurassic events in South America, Geología Colombiana, Bogotá, 20: 150-155. Toussaint, J. F., 1995b. Evolución geológica de Colombia 2. Triásico Jurásico. Contribución al IGCP 322 “Correlation of Jurassic events in South America” International Geological Correlation Programme Unesco IUGS. Univ. Nacional de Colombia. Medellín, 94 p. Toussaint, J. F. and Restrepo, J. J., 1974a. Algunas consideraciones sobre la evolución estructural de los Andes Colombianos. Informe parcial. Publicación Geológica Especial Geol., Univ. Nacional de Colombia, Medellín, 4: 1-17. Toussaint, J. F. and Restrepo, J. J., 1974b. La Formación Abejorral y sus implicaciones sobre la evolución de la Cordillera Central de Colombia durante el Cretáceo. Anales de la Facultad de Minas, Medellín, 58: 13-29. Toussaint, J. F. and Restrepo, J. J., 1989. Acresiones sucesivas en Colombia; un nuevo modelo de evolución geológica. V Congreso Colombiano de Geología, Bucaramanga, I: 127-146. Toussaint, J. F. and Restrepo, J. J., 1991. El magmatismo en el marco de la evolución geotectónica de Colombia. Simposio sobre el magmatismo Andino y su marco tectónico. Programa Internacional de Correlación Geológica, UNESCO. Unión Internacional de Ciencias Geológicas. Manizales, Colombia, Julio 3-5 de 1991. Manizales, p. 135-150.
290
References
Toussaint, J. F. and Restrepo, J. J., 1994. The Colombian Andes during Cretaceous times. In: J.A. Salfity (ed.), Cretaceous tectonics of the Andes. Views Brunwick, Germany, p. 61-100. Tron, V. and Brun, J. P., 1991. Experiments on oblique rifting in brittle-ductile systems. Tectonophysics, 188: 71-84. Trygvason, E. and Lawson, J. E. Jr., 1970. The intermediate earthquake source near Bucaramanga, Colombia. Bull. Seism. Soc. Am., 60: 269-276. Turcotte, D. L. and Schubert, G., 1982 . Geodynamics, applications of continuum physics to geological problems, John Wiley New York., 450 p. Ujueta, G., 1961. Geología del Noreste de Bogotá. Boletín Geológico Servicio Geológico Nacional Bogotá, 9(1-3): 23-46. Ujueta, G., 1993. Lineamientos Muzo, Tunja y Paipa en los Departamentos de Boyacá y Casanare, Colombia. Geología Colombiana, 18: 65-73. Ulloa, C. and Rodríguez, E., 1976a. Geología del cuadrángulo K-12, Guateque. Boletín Geológico, Ingeominas, Bogotá, 22(1): 4-55. Ulloa, C. Rodríguez, E., et al., 1976b. Geological map Cuadrángulo K-13. Tauramena . Cuenca de los Farallones. Scale 1:100.000, Ingeominas, Bogotá. Ulloa, C. and Roddríguez, E., 1979a. Geología de las planchas 170 velez y 190 Chiquinquirá, Colombia. Ingeominas Informe 1794, Sogamoso, 45 p. Ulloa, C. and Rodríguez, E., 1979b. Geología del Cuadrángulo K-13, Tauramena. Boletin Geológico, Ingeominas, Bogotá, 24(2): 3-30. Ulloa, C. and Rodríguez, E., 1987. Mapa geológico de la plancha 190, chiquinquirá, Map Scale 1:100.000, ingeominas, Bogotá. Ulloa, C. and Rodríguez, E., 1991. Mapa geológico de Colombia, plancha 190-Chiquinquira (Departamento de Boyacá), Ingeominas, Santafe de Bogotá, Scale 1:100.000, 26 p. Ulloa, C., Rodríguez, E. and Guerra, A., 1973. Mapa geológico de la plancha 172 Paz de Río. Scale 1:100.000, Ingeominas Bogotá. Ulloa, C. Camacho, R. and Escovar, R., 1975. Mapa Geológico de la Plancha K-12 de Guateque: Ingeominas, Esc 1:100.000, Bogotá. Ulloa, C.; Escobar, R. and Pacheco, A., 1976c. Mapa Geológico de la Plancha 230 de Monterrey: Ingeominas, Esc 1:100.000, Bogotá. United States Geological Survey, National Earthquake Information Center. World Data Center for Seismology, (http://www.neic.cr.usgs.gov/neis/epic/epic.html). Ureta, J.E. and Du Toit, Ch., 1997. Prospección de hidrocarburos en el piedemonte de la Cuenca del Putumayo. VI Simposio Exploración petrolera en las Cuencas Subandinas. Cartagena de Indias, Mem. Tomo I: 263-273. Uyeda, S. and McCabe, R., 1983. A possible mechanism of episodic spreading of the Philippine Sea. In: M. Hashimoto and S. Uyeda, (eds.), Accretion tectonics in the Circum-Pacific regions. Terra, Tokyo, p. 291-306. Van Andel, T. H., 1958. Origin and classification of Cretaceous, Paleocene and Eocene sandstones of western Venezuela. A.A.P.G. Bull., 42(4): 734-763. Van der Beek, P., 1995. Tectonic evolution of continental rifts, inferences from numerical modelling and fission track thermochronology. Ph.D. Thesis, Vrije Universitteit, Amsterdam, 232 p. Van der Beek, P. A. and Cloetingh, S., 1992. Lithospheric flexure and the tectonic evolution of the Betic Cordilleras (SE Spain). Tectonophysics, 203: 325-344. Van der Hammen, T., 1958. Estratigrafía del Terciario y Maestrichtiano continentales y tectogénesis de los Andes Colombianos. Boletin Geológico, Ingeominas, Bogotá, 6(1-3): 67-128. Van der Hamen T. H., 1961. Late Cretaceous and Tertiary stratigraphy and tectogenesis of the Colombian Andes. Geologie en Mijnbouw, 40: 181-188. Van der Hammen T. and Hooghiemstra H., 1997. Chronostratigraphy and correlation of the Pliocene and Quaternary of Colombia, Quaternary International, 40: 81-91. Van der Hammen, T., Werner, J. H. and Van Dommelen, H., 1973. Palynological record of the upheaval of the northern Andes: a study of Pliocene and lower Quaternary of the Colombian Eastern Cordillera and the early evolution of its high-Andean biota. Paleobotany and Palynology, 16: 1-122.
291
References
Van der Wiel, A. M., 1991. Uplift and volcanism of the SE Colombian Andes in relation to Neogene sedimentation of the Upper Magdlena Valley. Ph.D. Thesis, Wageningen, 208 p. Van der Wiel, A. M. and Andriessen, P. A. M., 1991. Precambrian to recent thermotectonic history of the Garzón Massif (Eastern Cordillera of the Colombian Andes) as revealed by fission track analysis. In: A.M. Van der Wiel., Uplift and volcanism of the SE Colombian Andes in relation to Neogene sedimentation of the Upper Magdalena Valley. Ph.D. Thesis, Wageningen, 208 p. Van der Wiel, A. M. and Van den Bergh., 1992a. Uplift, subsidence, and volcanism in the southern Neiva Basin, Colombia, Part 1: Influence on fluvial deposition in the Miocene Honda Formation. Jour. South Am. Earth Sciences, 5(2): 153-175. Van der Wiel, A. M. and Van den Bergh., 1992b. Uplift, subsidence, and volcanism in the southern Neiva Basin, Colombia, Part 2: Influence on fluvial deposition in the Miocene Gigante Formation. Jour. South Am. Earth Sciences, 5(2): 175-196. Van Houten, F. B., 1976. Late Cenozoic volcaniclastic deposits, Andean foredeep, Colombia. Geol. Soc. Am. Bull., 87(4): 481-495. Van Houten, F. B. and Travis, R. B., 1968. Cenozoic deposits, Upper Magdalena Valley, Colombia. A.A.P.G. Bull., 52: 675-702. Van Wees, J. D., 1993. Flex3D, version 1.01, Dec. 1993. User manual. Netherlands Research School of Sedimentary Geology, Faculty of Earth Sciences, Vrije Universiteit, Amsterdam, 13 p. Van Wees, J. D., 1994. Tectonic modelling of basin deformation and inversion dynamics. The role of pre-existing faults and continental lithosphere rheology in basin evolution. PhD thesis Vrije Universiteit, Amsterdam, 164 p. Van Wees, J. D., 2000. Basin deformation partitioning and the role of pre-existing weak zones. Geophysical Research Abstracts, 25 th General Assembly European Geophysical Society. Abstract in Volume 2, Compact Disc. Van Wees, J. D. and Cloetingh, S., 1993. Role of pre-existing weak zones during basin extension and inversion. Terra Abstracts. J. Suppl. 1, p. 211. Van Wees, J. D. and Cloetingh, S., 1994. A finite difference technique to incorporate spatial variations in rigidity and planar faults into 3-D models for lithospheric flexure. Geoph. Jour. Int. 117(1): 179-195. Van Wees, J. D. and Stephenson, R. A., 1995. Quantitative modelling of basin and rheological evolution of the Iberian Basin (central Spain); implications for lithospheric dynamics of intraplate extension and inversion. In: S. Cloetingh, D. Argenio, B., Catalano, R., Horvath, F., Sassi, W. (eds.), Interplay of extension and compression in basin formation. Tectonophysics, 252 (1-4): 163-178. Van Wees, J. D. and Beekman, F., 2000. Lithosphere rheology during intraplate basin extension and inversion – Inferences from automated modelling of four basins in western Europe. Tectonophysics, 320: 219-242. Van Wees, J. D., De Jong, K. and Cloetingh, S., 1992. Two dimensional modeling and the dynamics of extension and inversion in the Betic zone (SE Spain). Tectonophysics, 203: 305-324. Van Wees, J. D., Cloetingh, S. and De-Vicente, G., 1996a. The role of pre-existing faults in basin evolution; constraints from 2-D finite element and 3-D flexural models. In: P. G. Buchanan and D. A. Nieuwland, (eds.), Modern developments in structural interpretation valildation and modelling . Geol. Soc. Spec. Publ. London, 99: 297-320. Van Wees, J. D., Stephenson, R. A., Stovba, S. M. and Shymanovskyi, V. A., 1996b. Tectonic variation in the Dnieper-Donets Basin from automated modelling of backstripped subsidence curves. Tectonophysics, 268(1-4): 257-280. Van Wees, J. D., Arche, A., Beijdorff, C. G., López-Gómez, J. and Cloetinhg, S., 1998. Temporal and spatial variations in tectonic subsidence in the Iberian Basin (eastern Spain): inferences from automated forward modelling of high resolution stratigraphy (Permian-Mesozoic). Tectonophysics, 300:285-310. Vanegas, D. and Arango, F., 1994. Análisis estratigráfico de una región al Occidente de la población de Paz de Ariporo (Casanare). Tesis Pregrado, Univ. Nacional de Colombia, Bogotá, 70 p. Vargas, R. and Arias, A., 1981a. Mapa geológico de la plancha 86, Abrego, Scale 1:100.000, Ingeominas, Bogotá.
292
References
Vargas, R. and Arias, A., 1981b. Mapa geológico de la plancha 97, Cáchira, Scale 1:100.000, Ingeominas, Bogotá. Vargas, R., Arias, L., Jaramillo, L. and Tellez, N., 1984a. Geología del cuadrángulo I-13, Malaga, Boletín Geológico Ingeominas, Bogotá, 24(3): 1.76. Vargas, R., Arias, L., Jaramillo, L. and Tellez, N., 1984b. Mapa geológico preliminar de la plancha 136, Malaga, Scale 1:100.000, Ingeominas, Bogotá. Vargas, R., Etayo, Renzoni, G. and Téllez, N., 1985. Corte estratigráfico panorámico de la Formación Los Santos, Carretera Quebrada El Medio El Boquerón Santander. In: F. Etayo-Serna and F. Laverde-Montaño (eds.), Proyecto Crétacico, contribuciones. Chapter V, Ingeominas Publicación Geológica Especial 16, Bogotá, 12 p. Vargas, R.; Arias, A.; Jaramillo, L and Tellez, N., 1987. Mapa Geológico de la Plancha 152 de Soata: Ingeominas, Esc 1:100.000. Vásquez, C. A., 1983. Geología del Paleoceno Superior en la margen occidental de la Cuenca de los Llanos Orientales. Tesis Pregrado Geología Univ. Nacional de Colombia, Bogotá, 91 p. Vásquez, C. A., 1988. Mapa estructural del tope del Cretáceo, Cuenca Llanos Orientales de Colombia. III Simposio Bolivariano Exploración Petrolera en las Cuencas Subandinas, Caracas, 1: 580609. Vásquez, L. E. and Ríos, O., 1979. Geología y geoquímica de sedimentos activos de la región situada al norte de dolores (Tolima). Tesis pregrado Univ. Nacional de Colombia, Depto. Geociencias, Bogotá, 80 p. Vergara, L. and Prössl , K. F., 1994. Dating the Yaví Formation (Aptian Upper Magdalena Valley, Colombia). In: F. Etayo Serna (ed.), Estudios geológicos del Valle Superior del Magdalena. Chapter XVIII, Univ. Nacional de Colombia, Ecopetrol, Bogotá, 14 p. Villamil, T.,1993. Relative sea level, chronology, and a new sequence stratigraphy model for distal offshore facies, Albian to Santonian, Colombia. In Pindell, J. A. and Drake, C. D. (eds), Mesozoic-Cenozoic stratigraphy and tectonic evolution of the Caribbean Region/Northern South America: Implications for eustasy from exposed sections of a Cretaceous-Eocene passive margin setting. Geol. Soc. Am. Memoir, paper C-8. Villamil, T., 1994. High-resolultion stratigraphy, chronology and relaitve sea level of the AlbiianSantonian (Cretaceous) of Colombia. Ph.D. Thesis Univ. Of Colorado at Boulder, 446 p. Villamil, T., 1999. Campanian-Miocene tectonostratigraphy, depocenter evolution and basin development of Colombia and western Venezuela. Palaeogeography, Palaeoclimatology, Palaeoecology, 153(1-4): 239-275. Villamil, T. and Restrepo, P., 1997. Paleocene-Miocene paleogeographic evolution of Colombia. Memorias VI Simposio Bolivariano “Exploración Petrolera en las Cuencas Subandinas”., Cartagena, Sept.14-17, 1997. Tomo I: 275-302. Villamil, T. and Arango, C., 1998. Integrated stratigraphy of latest Cenomanian and early Turonian facies of Colombia. Paleogeographic evolution and non-glacial eustasy, Northern South America, SEPM Spec. Publ. 58: 129-159. Villamil, T. and Pindell, J. L., 1998. Mesozoic paleogeographic evolution of northern South America: Foundations for sequence stratigraphy studies in passive margin strata deposited during nonglacial times. Paleogeographic evolution and non-glacial eustasy, Northern South America, SEMP. Spec. Publ. 58: 283-318. Villarroel, C. and Guerrero, J., 1984. Un nuevo y singular representante de la familia leontiniidase? (notoungulata, Mammalia) en el Mioceno de La Venta, Colombia. Geología Norandina, Bogotá, 9: 35-40. Waddell, M., 1982. Depositional model for the Monserrate Formation of the Neiva Basin, Colombia, South America. MS. Thesis, Univ. South Carolina, Columbia, SC. 83 p. Walcott, R. I., 1970. Flexural rigidity, thickness and viscosity of the lithosphere. J. Geophys. Res. , 75: 3941-3955. Walthall, B. H. and Berry, D. W., 1979. Paipa-Belencito-Paz de Río area. In Geotec (Ed.), Geological Field Trips, Colombia. Colombian Society of Petroleum Geologists and Geophysicists., Ediciones Geotec, Bogotá,: 65-87.
293
References
Ward, D., Golsmith, R., Cruz, J. and Restrepo, H., 1973. Geología de los cuadrángulos H-12, Bucaramanga y H-13, Pamplona, Depto. de Santander. Boletín Geológico Ingeominas, Bogotá, 21 (1-3): 1-132. Watts, A. B., Karner, G. D. and Steckler, M. S., 1982. Lithospheric flexure and the evolution of sedimentary basins. Phil. Trans. Roy. Soc. London, 305: 249-281. Wellman, S. S., 1967. Stratigraphy, petrology and sedimentology of the nonmarine Honda Formation (Miocene), Upper Magdalena Valley, Colombia. Ph.D. Thesis, Princeton Univ., Princeton, 250 p. Wellman, S. S., 1970. Stratigraphy and petrology of the nonmarine Honda Group (Miocene), Upper Magdalena Valley, Colombia. Geol. Soc. Am. Bull., 81:2358-2374. Wiggins, R. A., 1972. The general linear inverse problem: Implication of surface waves and free occillations for earth structure. J. Geophys. Res. , 80: 1053-1064. Wijninga, V. M. and Kuhry P., 1993. Late Piocene paleoecology of the Guasca Valley (Cordillera Oriental, Colombia). Rev. Paleobotany and Palynology 78(1-2): 69-127. Wilson, M., 1993. Magmatism and the geodynamics of basin formation. Sediment. Geol., 86: 5-29. Wilson, M. and Guiraud, R., 1992. Magmatism and rifting in western and central Africa, from Late Jurassic to recent time. In: P. Ziegler (ed.), Geodynamics of rifting Volume II Case history studies, North, South America and Africa. Tectonophysics, 213(1-2): 203-225. Willingshofer, E., 2000. Extension in collisional orogenic belts: the Late Cretaceous evolution of the Alps and Carpathians. PhD thesis Vrije Universiteit, Amsterdam, 146 p. Woerner, G., Gerdes, A., Henk, A., Tanner, D., and Finger, F., 1999. Magmatism of thickened continental crust; examples from the Variscan Orogen (Southern Bohemian Massif) and Central Andes. In: P. Dietrich, W. Frank and B. J. Merkel, 89 th annual meeting of the Geologische Vereinigung: Old crust, new problems, geeodynamics and utilization. Abstracts and programme. Orogenic processes, quantification and modeling in the Variscan Belt. Terra Nostra, Bonn, 99(1): 212-213. Wortel, R. and Cloetingh, S., 1981. On the origin of the Cocos-Nazca spreading center. Geology, 9: 425-430. Wortel, M. J. R., and Spakman, W., 1992, Structure and dynamics of subducted lithosphere in the Mediterranean region. Proc. Royal Netherlands Academy of Sciences, 95: 325-347. Yurewicz, D. A., Advocate, D. M., Lo, H. B. and Hernandez, E. A., 1998. Source rocks and oil families, southwest Maracaibo Basin (Catatumbo sub-basin), Colombia. A.A.P.G. Bull, 82(7): 1329-1352. Zamarreño de Julivert, I., 1963. Estudio petrográfico de las calizas de la Formación Rosablanca en la región de la Mesa de Los Santos (Cordillera Oriental, Colombia). Boletín Geológico Univ. Industrial de Santander, Bucaramanga, 15: 5-34. Ziegler, P. A., 1988. Evolution of the Artic-North Atlantic and Western Tethys. A.A.P.G., Mem., 43, 198p. Ziegler, P. A., 1990. Geological atlas of Western and Central Europe. 2nd ed. Shell Internationale Petroleum Mij. B.V., Geol. Soc., London, 239 p. Ziegler, P. A.,1993. Plate-moving mechanisms: their relative importance. Jour. Geol. Soc., London 150: 927-940. Ziegler, P. A.,1994. Geodynamic processes governing development of rifted basins. In: Roure, F., Ellouz, N., Shein, V. S. and Skvortsov, I. (eds.), Geodynamic evolution of sedimentary basins. Institut Français du Pétrole, Technip, Paris, p. 16-97. Ziegler, P. A., Cloetingh, S. and Van Wees, J. D., 1995. Dynamics of intra-plate compressional deformation: The Alpine foreland and other examples. Tectonophysics, 252: 7-59. Ziegler, P. A., Van Wees, J. D. and Cloetingh, S., 1998. Mechanical controls on collision-related compressional intraplate deformation. Tectonophysics, 300: 103-129. Zoetemeijer, R., 1993. Tectonic modelling of foreland basins. Thin skinned thrusting, syntectonic sedimentation and lithosphere flexure. Ph.D. Thesis, Vrije Universiteit, Amsterdam, 148 p. Zoetemeijer, R., 1998. Flexure and gravity modelling of foreland basins. User manual Cobra version 5. Netherlands Research School of Sedimentary Geology, Faculty of Earth Sciences, Vrije Universiteit, Amsterdam, 17 p.
294
